General psychopathology as a potential mediator between childhood trauma and psychosis

(1)

Graduate School of Psychology

Research Master’s Psychology Internship Report

Date: August / 2015

1. WHO AND WHERE Student

Name : Adela-Maria Isvoranu

Student ID number : 10749276

Address : Weesperplein 4

Postal code and residence : 1018XA Amsterdam Telephone number : +316 25220808

Email address : isvoranu.adela@gmail.com

Supervisor(s)

Within ResMas (obligatory) : Prof. Denny Borsboom, Drs. Claudia van Borkulo Specialisation : Psychological Methods & Clinical Psychology External supervisor(s), if any : Dr. Lindy-Lou Boyette

Research center / location : University of Amsterdam / Amsterdam Medical Centre

(2)

Internship Report

General Psychopathology as a Potential Mediator between Childhood Trauma and Psychosis

Adela M. Isvoranu

(3)

Abstract

Psychosis is a chronic, severe mental disorder, and among the top causes of long-term disability. Early trauma has been identified as a potential risk factor for psychosis, but to date there is little consensus in respect to which specific symptoms may be triggered by trauma, and what pathways may account for these associations. The purpose of the present study is to use a novel approach – the network approach – to investigate how traumatic childhood experiences relate to individual symptoms of the disorder, and identify likely pathways from trauma to psychosis. In networks, symptoms are viewed as causally connected, mutually reinforcing, and can have different attributes. To construct the networks, we used data (n=552) from the longitudinal observational study Genetic Risk and Outcome of Psychosis Project (GROUP), and included all symptom dimensions of the Positive and Negative Syndrome Scale (PANSS), and the five scales of the Childhood Trauma Questionnaire – Short Form (CTQ-SF). Our results suggest that general psychopathology may be a strong mediator between trauma and psychotic symptoms. Furthermore, sexual abuse and emotional neglect exhibited no direct associations with any of the symptoms. The findings partially support the existence of an affective pathway to psychosis, and raise the question if the detrimental effect of trauma is indeed specific to psychosis, or if it is a shared factor for major psychiatric disorders.

(4)

Introduction

Psychosis is a highly heterogeneous mental disorder that represents a leading cause of long-term disability, and can result in premature mortality as well as severe social and occupational impairment (Stafford, Jackson, Mayo-Wilson, Morrison, & Kendall, 2013). Childhood trauma (CT) has been extensively investigated as a potential risk factor for the onset of psychosis, and found to relate to some of the most severe forms of symptomatology in adulthood (e.g., Larkin & Read, 2008; Read, Agar, Argyle, & Aderhold; Spence et al., 2006). Children who had experienced early trauma face an increased risk for developing psychosis, up to 33% higher than the estimated population risk (Varese et al., 2012). The present study aims to contribute to the efforts of revealing the nature of the relationship between CT and psychosis by applying a novel approach to psychopathology - the network approach - which conceptualizes mental disorders as a causal system of interacting symptoms (Borsboom & Cramer, 2013).

The network approach has emerged in response to the poor outcomes of contemporary analyses carried out in psychopathology, which are based on the common cause hypotheses (Schmittmann et al., 2013), and conceptualize mental disorders as reflective latent variables that cause a wide array of symptoms. In other words, symptoms are taken to be indicators of an underlying disease entity, and high correlations between symptoms can be fully explained by the common influence of the latent disorder. Despite decades of research into mental disorders, however, finding such underlying causes for symptoms has been very rare. Instead, the causes appear to be multifactorial, thus challenging the likelihood of a common cause explanation for associations between symptoms (McNally et al., 2014). In network models, correlations between symptoms are no longer explained by the common latent factor (i.e., mental disorder), but symptoms are assumed to have autonomous causal power to influence each other (Cramer, Waldorp, van der Maas, & Borsboom, 2010). As such, they are no longer regarded as measurement of an underlying disorder, but causal associations among symptoms

(5)

are conceptualized to constitute the disorder. For instance, an external trigger could produce a certain symptom (e.g., anxiety), which could in turn activate other symptoms (e.g., tension). In recent years, the network approach has contributed to several advancements in psychopathology research (e.g., Fried et al., 2015; McNally et al., 2014, Ruzzano, Borsboom, & Geurts, 2015) and personality research (e.g., Cramer et al., 2012).

The current paper aims to address two issues. First, while there are a number of reports suggesting that CT only evokes specific symptoms rather than all symptoms to the same degree, there is little consensus in respect to this matter. Psychotic disorders are defined by abnormalities in one or more of the following domains: delusions, hallucinations, disorganized thinking, grossly disorganized abnormal motor behavior, and negative symptoms (American Psychiatric Association, 2013). It has been argued that different types of CT are likely to differ in the symptoms they are associated with (Bentall, Wickham, Shevlin, & Varese, 2012). For instance, Whitfield, Dube, Felitti, & Anda (2005) found, in a large population-based study (n=17337) that respondents showed a significant and graded relationship between CT - particularly physical and sexual abuse - and hallucinations. Other studies reported a significant relationship between CT and both hallucinations and delusions; Read and Argyle (1999) identified, in a clinical sample, that hallucinations may be more common among patients who have been sexually abused during childhood, while delusions may be more common among the physically abused patients. Correspondingly, Janssen and colleagues (2004) reported that early CT predicted the development of both hallucinations and delusions, and this association remained after adjusting for demographic and clinical variables. In contrast, Read et al. (2003) identified child abuse as a significant predictor of auditory and tactile hallucinations, while only the combination of child abuse and adult abuse was found to predict both hallucinations and delusions. In a more recent study that controlled for the co-occurrence of hallucinations and delusions, Bentall et al. (2012) found that childhood rape was only associated with hallucinations, while institutional care was

(6)

associated with paranoid delusions; both hallucinations and paranoid delusions were found to be more common among physically abused patients.

Second, the exploration of potential mechanistic pathways that account for the association between CT and psychosis is at very best in its infancy. It has been hypothesized that traumatic experiences may result in structural and neurochemical abnormalities in the brain and nervous system, affecting the function of the hypothalamus-pituitary-adrenal axis, which plays a role in the stress response (Bremner, 2002; Read et al., 2005). As such, if early trauma leads to an enduring sensitivity to daily life stress, it may be possible that CT may impact on the affective pathway to psychosis by increasing stress-sensitivity. Alternatively, cognitive models of psychosis argue that trauma may lead to negative beliefs about the self, world, and others, and these beliefs may in turn lead to distressing interpretations of everyday events, eventually resulting in psychotic experiences (Birchwood, Meaden, Trower, Gilbert, & Plaistow, 2000; Morrsion, Frame, & Larking, 2003).

One limitation to current studies is that these investigated the relationship between positive psychotic symptoms (primarily hallucinations and delusions) and CT, but they overlooked other symptom scales. To date, the most widely used questionnaire to assess psychotic symptom severity is the Positive and Negative Syndrome Scale (PANSS; Kay et al., 1987). The PANSS consists of three subscales: the positive scale (e.g., hallucinations, paranoia), the negative scale (e.g., social withdrawal, blunted affect), and the general psychopathology scale (e.g., anxiety, depression). In the present study, all three dimensions of the PANSS will be included, with the purpose of gaining insights into both the relationship between CT and psychotic symptoms, and the potential pathways that account for this association.

(7)

Method Participants

We analyzed data from the longitudinal observational study “Genetic Risk and Outcome of Psychosis Project” (GROUP; Korver et al., 2012) to examine the relationship between CT and psychotic disorders. The full sample consists of psychotic patients (n=1120), their siblings (n=1057), their parents (n=919), and a control group (n=590); in the present study, we used data from the patient sample only. The patients were recruited from 36 Mental Health Care institutions from four Academic Medical Centers (Amsterdam, Groningen, Maastricht, Utrecht). They were aged between 16 and 50 years, they met criteria for a non-affective psychotic disorder (schizophrenia, schizophreniform disorder, schizo-affective disorder, delusional disorder, psychotic disorder not otherwise specified), and the estimated level of intelligence was above 70. For further information about the sample, please refer to Korver et al. (2012).

Symptomatology

We used the first wave of data from the Positive and Negative Syndrome Scale (PANSS; Kay et al., 1987) as a measure of symptom severity in the patient population. The PANSS consists of 30 items divided in three subscales: the positive scale (e.g., hallucinations, paranoia), the negative scale (e.g., social withdrawal, blunted affect), and the general psychopathology scale (e.g., anxiety, depression). It is scored on a seven-point Likert-type scale, ranging from 1 (absent) to 7 (very severe).

Childhood Trauma

Childhood trauma was measured with the Dutch version of the Childhood Trauma Questionnaire Short Form (CTQ-SF; Bernstein et al., 2003; Thombs et al., 2009). The data were collected at baseline (first wave of data collection) at the Maastricht site, while the other

(8)

sites collected it at the follow-up measurement (second wave of data collection). The CTQ-SF is a self-report questionnaire, consisting of 24 items, and it is scored on a five-point Likert-type scale, ranging from 1 (never true) to 5 (always true). The CTQ measures: physical neglect (failure of caretaker to provide basic necessities for a child such as food, clothing, shelter); physical abuse (bodily assault on a child posing a risk of or resulting in injury); emotional neglect (failure of caretaker basic emotional and psychological needs for a child, such as love and nurturance); emotional abuse (verbal assaults on a child, such as humiliation); and sexual abuse (unwanted sexual contact or conduct between a child and an adult). Each scale encompassed five items, with the exception of the sexual abuse scale, from which the item “Molestation” was removed due to improper translation into Dutch. A sum-score was calculated for each scale and used when computing the networks – the sum of the sexual abuse scale was multiplied by 5/4 to have the same mean levels as the other scales.

Network Construction

To visualize the relationship between CT and psychotic disorders, symptom networks were constructed. In networks, individual items (i.e., symptoms) are represented as nodes, and the connection between two nodes (edge) depicts a direct influence between them (Borsboom and Cramer, 2013). Each node in the network represents one of the thirty symptoms as included in the PANSS questionnaire, and the remaining five nodes represent the five scales of the CTQ-SF. To construct the network, a Gaussian Graphical Model (Lauritzen & Wermuth, 1989) was estimated, in which a connection between two nodes denotes a partial correlation between the two, while controlling for all other nodes. The shrinkage parameter graphical lasso (glasso; Friedman, Hastie, & Tibshirani, 2008) was used to control for potential spurious connections in the network. The resulting network was further analyzed by investigating the importance (centrality) of each node in the network. This can be captured in three centrality measures: node strength, betweenness, and closeness (Barrat, 2004;

(9)

Boccaletti, 2006; Opsahl, 2010). In weighted networks, node strength is a measure of the number and strength of connections. Betweenness measures how often a node lies on the shortest path between every combination of two other nodes, indicating to what extent the node facilitates the flow of information though the network. The closeness of a node measures the average distance from that node to all other nodes in the network, a high closeness indicating a short average distance between a given node and the remaining nodes in the network.

Following the symptom networks, networks illustrating the shortest paths between each trauma scale and the positive and negative symptoms of the PANSS were computed. The shortest path between two nodes represents the steps one needs to take to reach one node from the other (e.g., to reach node “D” from node “A”, one needs to go through nodes “B” and “C”; Brandes, 2008), and it is computed using Dijkstra’s algorithm (Dijkstra, 1959). Our networks illustrate which are the shortest routes from each CT scale to the different clusters of positive and negative psychotic symptoms. In this way, we can see possible mediating items between trauma and psychosis.

All analyses were performed using the R-statistical software (R Core Team, 2013). The networks were constructed and visualized using the R-package qgraph (Epskamp, Cramer, Waldorp, Schmittmann, & Borsboom, 2012). The default layout algorithm in qgraph is Fruchterman and Reingold’s (1991) layout, which places the nodes with stronger connections into the center of the network, and the nodes with weaker connections closer to the periphery of the network.

Results

In total, after removing all missing data, 552 patients were included in the analytic sample. The patients, of whom 75% were males, had a mean age of 30.8 years (SD = 7.27). Table 1 presents the demographic and clinical characteristics of the sample, the mean sum-scores on

(10)

the three dimensions of the PANSS, and the five dimensions of the CTQ-SF; Table 2 presents the item distribution and the item labels for all following networks. The current data were not univariate normally distributed, and as such, a non-paranormal transformation to relax the normality assumption was applied prior to constructing the networks, as suggested by Liu, Lafferty & Wasserman (2009).

In sum, we found that the general psychopathology scale of the PANSS mediates the relationship between trauma and psychosis. To better illustrate our results, we will first present a network that only includes the CT scales and the positive and negative symptoms of the PANSS, and then we will describe the network that includes the CT scales and all the three symptom dimensions of the PANSS.

#INSERT TABLE 1 ABOUT HERE#

Childhood Trauma and PANSS

The first two networks we constructed illustrate the relationship between CT and (1) only the first two scales of the PANSS (i.e., positive symptoms scale, negative symptoms scale), and (2) all three scales of the PANSS. Each edge within the network corresponds to a partial correlation between two individual items. The thickness of an edge represents the absolute magnitude of the correlation (the thicker the edge, the stronger the connection), while the color of the edge indicates the size of the correlation (green for positive connections, red for negative connections; Epskamp et al., 2012).

#INSERT FIGURE 1 ABOUT HERE#

When evaluating the first network (Figure 1), it can be observed that no connection between any of the CT scales and negative psychotic symptoms is present. However, nodes

(11)

CT5 (emotional abuse), CT4 (physical abuse), and CT3 (sexual abuse) are directly associated with two of the positive psychotic symptoms: P6 (paranoia) and P7 (hostility); Nodes CT3 and CT4 have a direct positive edge to item P6, while node CT5 has a direct positive edge to item P7. In addition, all items are highly interconnected within their given scale, implying that correlations among subscales of one scale are much larger than correlations across scales. Centrality measures of the network (see Supplementary Materials) show that items P2 (conceptual disorganization) and N7 (stereotyped thinking) have the highest closeness and strength measures, while item P6 (paranoia) has high betweenness and closeness measures. This can also be noticed within the network, as items P2 and N7 are positioned closer to the center of the network, while item P6 is the main node that facilitates the connection between trauma and psychotic symptoms.

The second network (Figure 2) includes all the three scales of the PANSS. Strikingly, within this network, there is no connectivity between the CT scales and positive or negative psychotic symptoms; CT only connects to the general psychopathology scale of the PANSS. Node CT5 (physical abuse) is positively associated with GP1 (somatic concern) and GP14 (poor impulse control), and negatively associated with GP5 (mannerism and posturing). Node CT4 (emotional abuse) is positively associated with item GP2 (anxiety), node CT3 (sexual abuse) is positively associated with item GP3 (guilt), and node CT1 (physical neglect) is positively associated with item GP7 (motor retardation). Centrality measures of the network (see Supplementary Materials) show that item GP9 (unusual thought concern) now has the highest betweenness, closeness, and strength measures; nonetheless, items P2 (conceptual disorganization), N7 (stereotyped thinking), and P6 (paranoia) are still central to the network.

(12)

#INSERT TABLE 2 ABOUT HERE#

Shortest Paths: Childhood trauma - Positive and Negative Psychotic Symptoms

Following the symptom networks, we constructed five networks that depict shortest paths between each trauma scale and all individual symptoms of the PANSS. Due to high similarities between these networks and space constraints, only two such networks will be described in this section, and the rest can be accessed in Supplementary Materials.

Networks three and four illustrate the shortest paths between the node CT3 (sexual abuse) and the positive and negative symptoms (Figure 3a), and the node CT5 (physical abuse) and the positive and negative symptoms (Figure 3b). In other words, the networks display the shortest routes that connect the nodes CT3 and CT5 to each individual positive and negative symptom of the PANSS. These two specific CT scales were chosen because they are distinct from each other, and they illustrate nearly all possible routes from other trauma scales to the psychotic symptoms as well.

In Figure 3a, the shortest route to reach most negative psychotic symptoms from node CT3 (sexual abuse) is via node CT1 (physical neglect) and GP7 (motor retardation). The activation to the items P1 (delusions), P3 (hallucinations), P6 (paranoia) spreads via items CT4 (emotional abuse) and GP2 (anxiety), while the activation to the remaining positive items P2 (conceptual disorganization), P4 (hyperactivity), P5 (grandiosity), and P7 (hostility) spreads via node CT5 (physical abuse) and item GP14 (poor impulse control). Lastly, to reach item N7 (stereotyped thinking) from CT3, the shortest route is via node CT5 (physical abuse), GP14 (poor impulse control), and GP15 (preoccupation). In sum, node CT3 (sexual abuse) does not appear to directly connect to any of the PANSS symptoms, but only to the other trauma scales (i.e., sexual abuse leads to physical neglect, physical abuse, and emotional abuse), which then activate the rest of the network.

(13)

retardation), which then triggers the negative symptoms N1 (blunted affect), N2 (emotional withdrawal), N3 (poor rapport), N4 (social withdrawal) and N6 (lack of spontaneity and flow of conversation). Node CT5 appears to elicit positive symptoms through three pathways: (1) CT5 activates the items GP1 (somatic concern) and GP9 (unusual thought concern), which then triggers the items P1 (delusions) and P3 (hallucinations); (2) CT5 activates item GP14 (poor impulse control), which then triggers items P2 (conceptual disorganization), P4 (hyperactivity), P5 (grandiosity), and P7 (hostility); (3) CT5 activates the trauma node CT4, which then triggers item GP2 (anxiety), followed by item P6 (paranoia).

The networks depicting the shortest paths between the remaining three CT scales - CT1 (physical neglect), CT2 (emotional neglect), and CT4 (emotional abuse) - and the positive and can be accessed in Supplementary Materials; these illustrate a similar pattern of activation as the two networks described above: the activation to all positive and negative psychotic symptoms first spreads (directly) via the general psychopathology items, or first via another CT scale, followed by one or more of the general psychopathology items. For the nodes CT2 (emotional neglect) and CT3 (sexual abuse) all connectivity always first spreads via another CT node. For the node CT4 (emotional abuse), there is an activation route to items P1 (delusion), P3 (hallucinations), P6 (paranoia) via item GP2 (anxiety). In all the shortest path networks, node CT1 (physical neglect) is the only CT node that links trauma to most negative symptoms, via motor retardation.

Discussion

The present paper has provided the first network-based analysis of the relationship between traumatic childhood experiences and psychosis. In contrast to previous literature (e.g., Janssen et al., 2004; Whitfield et al., 2005), we did not identify a direct relation between

(14)

CT and hallucinations or delusions. In a network including only the positive and negative psychotic symptoms, we found that trauma is associated with most positive and negative symptoms only through paranoia and hostility. Furthermore, neither emotional neglect nor physical neglect had any direct association to any of the two psychotic symptom dimensions, but showed high connectivity to the three CT scales measuring abuse - in this context, it appears that neglect itself, in the absence of abuse, is not a trigger of psychotic symptoms. In the next network that included all three PANSS scales (i.e., the general psychopathology scale was added), trauma items were no longer associated with any of the positive or negative symptoms, but only to general psychopathology symptoms. In other words, when the general psychopathology scale was included in the analysis, the few direct associations identified between trauma and the positive and negative symptoms disappear. Thus, our results suggest that general psychopathology may be a mediator between trauma and psychosis. By computing shortest path networks, we illustrated which general psychopathology symptoms are activating different clusters of positive and negative symptoms.

Research Implications

Our findings have important research implications. First, the present paper has included in the analysis a wide array of symptoms, which are routinely measured in clinical practice (Opler & Ramirez, 1998). In comparison with previous research that investigated mainly hallucinations and delusions in relation to early trauma, examining the positive, negative, and general psychopathology symptoms yielded distinct and novel outcomes. The results of our study suggest that general psychopathology symptoms are mediating the relationship between trauma and psychosis. The question arises whether the effect of early trauma is specific to psychotic symptoms or rather it is specific to a wide array of symptoms that are present in most mental conditions. Research has shown patients suffering from anxiety, depression, and bipolar disorder often undergo traumatic experiences in childhood (e.g., Leverich et al., 2002;

(15)

Young, Abelson, Curtis, & Nesse, 1997); the PANSS’s general psychopathology scale measures symptoms that can also often be identified in several other mental disorders. This shows that early trauma may indeed be specific to a wide array of symptoms. Furthermore, psychiatric comorbidity is very common in psychotic disorders, often being present in up to 70% of the patients (Strakowski, Keck Jr., McElroy, Lonczak, & West, 1995), and this is not always accounted for. In our sample, comorbidity rates with depression and bipolar disorder were low, but not fulfilling the diagnostic criteria does not imply no disorder specific symptoms were present. In another sample where depression was accounted for, the relationship between sexual abuse and psychosis was found to be partly mediated by depression (Bebbington et al., 2011). Further research into the association between trauma and the subsequent development of psychopathology is warranted.

Second, even though research has so far been successful in establishing a connection between trauma and psychotic symptoms, there is little consensus in regard to potential mechanistic pathways that may account for this relation. Our results support in part the existence of an affective pathway to psychosis (Myin-Germeys & van Os, 2007) – trauma may lead to psychosis through a pathway of heightened emotional distress (e.g., depression, anxiety, guilt, tension); these symptoms indicative of emotional distress are some of the symptoms that comprise the general psychopathology scale of the PANSS. In line with this idea, Glaser and colleagues (2006) reported increased negative affect in subjects with a history of CT, which moderated the emotional reactivity to small daily stressors, providing further circumstantial evidence for an affective pathway to psychosis. Moreover, depression and anxiety were identified as strong predictors of psychotic symptoms (Owens et al., 2005), suggesting affective alterations in both disorders.

Several limitations of the current study should be taken into consideration. First, the majority of the sample studied in this report was male, and for males the rate of abuse is often lower than it is for females (e.g., Tolin & Foa, 2006). As such, a different pattern of results

(16)

may be observed for a more gender-balanced sample. Follow-up studies may investigate gender differences in relation to early trauma and psychosis. Second, in respect to the high number of symptoms included in the network, our sample can be considered a somewhat small sample; as a result, to account for potential high standard error rates, a minimum absolute value of .03 was used for edge weights when computing all networks. Third, the PANSS is a questionnaire measuring symptom severity, and not a questionnaire used in the diagnosis of psychosis – the four data collection sites used two distinct diagnosis questionnaires, and as such we were unable to use them in the analysis; our sample was composed of subjects already diagnosed with an affective psychotic disorder, and as such, using a symptom severity measure should not pose as a problem.

Conclusions

The goal of our study was to investigate the relationship between childhood traumatic experiences and psychotic symptoms by using an alternative approach to psychopathology - the network approach, which conceptualizes mental disorders as causal system of interacting symptoms. Our results suggest that general psychopathology may be a mediator between trauma and psychosis, providing evidence for an affective pathway to psychosis, and raising questions regarding the specificity of trauma to psychosis.

(17)

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author.

Barrat, A., Barthelemy, M., Pastor-Satorras, R., & Vespignani, A. (2004). The architecture of complex weighted networks. Proceedings of the National Academy of Sciences of the

United States of America, 101(11), 3747-3752.

Bebbington, P., Jonas, S., Kuipers, E., King, M., Cooper, C., Brugha, T., ... & Jenkins, R. (2011). Childhood sexual abuse and psychosis: data from a cross-sectional national psychiatric survey in England. The British Journal of Psychiatry, 199(1), 29-37.

Bentall, R. P., Wickham, S., Shevlin, M., & Varese, F. (2012). Do specific early-life adversities lead to specific symptoms of psychosis? A study from the 2007 the Adult Psychiatric Morbidity Survey. Schizophrenia bulletin, 38(4), 734-740.

Bernstein, D. P., Stein, J. A., Newcomb, M. D., Walker, E., Pogge, D., Ahluvalia, T., ... & Zule, W. (2003). Development and validation of a brief screening version of the Childhood Trauma Questionnaire. Child abuse & neglect, 27(2), 169-190.

Birchwood, M., Meaden, A., Trower, P., Gilbert, P., & Plaistow, J. (2000). The power and omnipotence of voices: Subordination and entrapment by voices and significant others. Psychological medicine, 30(02), 337-344.

Boccaletti, S., Latora, V., Moreno, Y., Chavez, M., & Hwang, D. U. (2006). Complex networks: Structure and dynamics. Physics reports, 424(4), 175-308.

Borsboom, D., & Cramer, A. O. (2013). Network analysis: an integrative approach to the structure of psychopathology. Annual review of clinical psychology, 9, 91-121.

Brandes, U. (2008). On variants of shortest-path betweenness centrality and their generic computation. Social Networks, 30(2), 136-145.

Bremner, J. D. (2002). Does stress damage the brain? Understanding trauma based disorders from a neurological perspective. New York: Norton.

(18)

Cramer, A. O. J., Sluis, S., Noordhof, A., Wichers, M., Geschwind, N., Aggen, S. H., ... Borsboom, D. (2012). Dimensions of normal personality as networks in search of equilibrium: You can’t like parties if you don't like people. European Journal of Personality, 26, 414–431.

Cramer, A. O., Waldorp, L. J., van der Maas, H. L., & Borsboom, D. (2010). Comorbidity: a network perspective. Behavioral and Brain Sciences, 33(2-3), 137-150.

Dijkstra, E. W. (1959). A note on two problems in connexion with graphs.Numerische mathematik, 1(1), 269-271.

Epskamp, S., Cramer, A. O., Waldorp, L. J., Schmittmann, V. D., & Borsboom, D. (2012). qgraph: Network visualizations of relationships in psychometric data.Journal of Statistical Software, 48(4), 1-18.

Fried, E. I., Bockting, C., Arjadi, R., Borsboom, D., Amshoff, M., Cramer, A. O., ... & Stroebe, M. (2015). From loss to loneliness: The relationship between bereavement and depressive symptoms. Journal of abnormal psychology,124(2), 256.

Friedman, J., Hastie, T., & Tibshirani, R. (2008). Sparse inverse covariance estimation with the graphical lasso. Biostatistics, 9, 432–441.

Fruchterman, T. M., & Reingold, E. M. (1991). Graph drawing by force-directed placement. Software: Practice and experience, 21(11), 1129-1164.

Glaser, J. P., Van Os, J., Portegijs, P. J., & Myin-Germeys, I. (2006). Childhood trauma and emotional reactivity to daily life stress in adult frequent attenders of general practitioners. Journal of psychosomatic research, 61(2), 229-236.

Janssen, I., Krabbendam, L., Bak, M., Hanssen, M., Vollebergh, W., Graaf, R. D., & Os, J. V. (2004). Childhood abuse as a risk factor for psychotic experiences. Acta Psychiatrica Scandinavica, 109(1), 38-45.

Kay, S., Fiszbein, A., & Opler, L. (1987). The Positive and Negative Syndrome Scale (PANSS) for Schizophrenia. Schizophrenia Bulletin, 13(2), 261-276.

(19)

Korver, N., Quee, P. J., Boos, H., Simons, C. J., & Haan, L. (2012). Genetic Risk and Outcome of Psychosis (GROUP), a multi site longitudinal cohort study focused on gene– environment interaction: objectives, sample characteristics, recruitment and assessment methods. International Journal of Methods in Psychiatric Research, 21(3), 205-221. Larkin, W., & Read, J. (2008). Childhood trauma and psychosis: evidence, pathways, and

implications. Journal of postgraduate medicine, 54(4), 287.

Lauritzen, S. L., & Wermuth, N. (1989). Graphical models for associations between variables, some of which are qualitative and some quantitative. The Annals of Statistics, 17, 31–57. Leverich, G. S., McElroy, S. L., Suppes, T., Keck, P. E., Denicoff, K. D., Nolen, W. A., ... &

Post, R. M. (2002). Early physical and sexual abuse associated with an adverse course of bipolar illness. Biological psychiatry, 51(4), 288-297.

Liu, H., Lafferty, J., & Wasserman, L. (2009). The nonparanormal: Semiparametric estimation of high dimensional undirected graphs. The Journal of Machine Learning Research, 10, 2295–2328.

McNally, R. J., Robinaugh, D. J., Wu, G. W., Wang, L., Deserno, M. K., & Borsboom, D. (2014). Mental Disorders as Causal Systems A Network Approach to Posttraumatic Stress Disorder. Clinical Psychological Science, 2167702614553230.

Morrison, A. P., Frame, L., & Larkin, W. (2003). Relationships between trauma and psychosis: a review and integration. British Journal of Clinical Psychology,42(4), 331-353.

Myin-Germeys, I., & van Os, J. (2007). Stress-reactivity in psychosis: evidence for an affective pathway to psychosis. Clinical psychology review, 27(4), 409-424.

Opler, L. A., & Ramirez, P. M. (1998). Use of the Positive and Negative Syndrome Scale (PANSS) in Clinical Practice. Journal of Psychiatric Practice, 4(3), 157-162.

Opsahl, T., Agneessens, F., & Skvoretz, J. (2010). Node centrality in weighted networks: Generalizing degree and shortest paths. Social Networks, 32(3), 245-251.

(20)

R Development Core Team. (2013). R: A Language and Environment for Statistical Computing. Vienna, Austria.

Read, J., & Argyle, N. (1999). Hallucinations, delusions, and thought disorder among adult psychiatric inpatients with a history of child abuse. Psychiatric Services, 50(11), 1467-1472.

Read, J., Agar, K., Argyle, N., & Aderhold, V. (2003). Sexual and physical abuse during childhood and adulthood as predictors of hallucinations, delusions and thought disorder. Psychology and psychotherapy, 76(1), 1-22.

Read, J., van Os, J., Morrison, A. P., & Ross, C. A. (2005). Childhood trauma, psychosis and schizophrenia: A literature review with theoretical and clinical implications. Acta Psychiatrica Scandinavica, 112, 330−350.

Ruzzano, L., Borsboom, D., & Geurts, H. M. (2015). Repetitive Behaviors in Autism and Obsessive–Compulsive Disorder: New Perspectives from a Network Analysis. Journal of autism and developmental disorders, 45(1), 192-202.

Schmittmann, V. D., Cramer, A. O., Waldorp, L. J., Epskamp, S., Kievit, R. A., & Borsboom, D. (2013). Deconstructing the construct: A network perspective on psychological phenomena. New Ideas in Psychology, 31(1), 43-53.

Spence, W., Mulholland, C., Lynch, G., McHugh, S., Dempster, M., & Shannon, C. (2006). Rates of childhood trauma in a sample of patients with schizophrenia as compared with a sample of patients with non-psychotic psychiatric diagnoses. Journal of trauma & dissociation, 7(3), 7-22.

Stafford, M. R., Jackson, H., Mayo-Wilson, E., Morrison, A. P., & Kendall, T. (2013). Early interventions to prevent psychosis: systematic review and meta-analysis. Bmj, 346, f185. Strakowski, S. M., Tohen, M., Stoll, A. L., Faedda, G. L., Mayer, P. V., Kolbrener, M. L., &

Goodwin, D. C. (1993). Comorbidity in psychosis at first hospitalization. American Journal of Psychiatry, 150, 752-752.

(21)

Thombs, B. D., Bernstein, D. P., Lobbestael, J., & Arntz, A. (2009). A validation study of the Dutch Childhood Trauma Questionnaire-Short Form: Factor structure, reliability, and known-groups validity. Child abuse & neglect, 33(8), 518-523.

Tolin, D. F., & Foa, E. B. (2006). Sex differences in trauma and posttraumatic stress disorder: a quantitative review of 25 years of research. Psychological bulletin, 132(6), 959.

Varese, F., Smeets, F., Drukker, M., Lieverse, R., Lataster, T., Viechtbauer, W., ... & Bentall, R. P. (2012). Childhood adversities increase the risk of psychosis: a meta-analysis of patient-control, prospective-and cross-sectional cohort studies. Schizophrenia bulletin, 38(4), 661-671.

Young, E. A., Abelson, J. L., Curtis, G. C., & Nesse, R. M. (1997). Childhood adversity and vulnerability to mood and anxiety disorders.

Whitfield, C. L., Dube, S. R., Felitti, V. J., & Anda, R. F. (2005). Adverse childhood experiences and hallucinations. Child abuse & neglect, 29(7), 797-810.

(22)

Figures

Figure 1. Network depicting two dimensions (positive and negative symptoms) of the

Positive and Negative Syndrome Scale (PANSS), and the five dimensions of the Childhood Trauma Questionnaire - Short Form (CTQ-SF). Symptom groups are differentiated by colors.

(23)

Figure 2. Network depicting the three dimensions (positive symptoms, negative symptoms,

general psychopathology) of the Positive and Negative Syndrome Scale (PANSS), and the five dimensions of the Childhood Trauma Questionnaire - Short Form (CTQ-SF). Symptom groups are differentiated by colors.

(24)

24 S h o rtes t P at h s S ex u al A bus e - P os iti ve a nd N ega tive S ym pt om s (b) S hor te st P at hs P hys ic al A bus e - P os iti ve a nd N ega tive S ym pt om s ur e 3 . ( a) N etw ork d ep ic ting sh ortes t p ath s b etw een th e S ex ual A bu se s cal e ( i.e, no de C T3 ) o f th e C hil dh oo d T rau m a Q ues tio nnair e-S ho rt F orm (C TQ -SF ), th e tw o m ain dim ens io n o f th e Po sit iv e and N eg ativ e Sy nd ro m e S cal e ( PA N SS ): p os iti ve s ym pto m s and neg ativ e s ym pto m s. ( b) N et w or k d ep ic ting ortes t p ath s b etw ee n th e Ph ys ic al A bu se s cal e ( i.e ., n od e C T5 ) o f t he C TQ -SF , and th e tw o m ain d im ens io ns o f th e P AN SS : p os itiv e s ym pto m s and n eg ativ e pto m s. D as hed and do tte d l ines rep res ent bac kg ro und c onnec tio ns ex is tent w ith in th e ne tw or k, th at ar e l es s r el ev ant w hen i nv es tig atin g s ho rtes t p ath s; ger d ots rep res ent s tro ng er co nnec tio ns , w hil e s m al ler d ots and das hed lines rep res ent w ea ker co nnec tio ns .

(25)

Table Captions

Table 1

Demographic and clinical characteristics: means (standard deviations) of the patient sample

Variable Men (n=418) Women (n=134) Total (n=552)

Age (years) 30.35 (6.59) 32.01 (8.99) 30.76 (7.27) Diagnostic % Schizophrenia 53.2 15.5 68.7 Schizophreniform Disorder 2.1 1.2 3.3 Schizoaffective Disorder 11.4 5.0 16.4 Delusional Disorder 1.2 0 1.2 Psychotic Disorder NOS 7.6 2.8 10.4 PANSS sum Positive symptoms 12.93 (5.42) 11.16 (4.64) 12.50 (5.29) Negative symptoms 13.65 (5.69) 12.04 (5.15) 13.26 (5.60) General Psychopathology 27.67 (8.26) 25.18 (7.72) 27.06 (8.20) CTQ-SF sum Emotional Neglect 18.80 (4.25) 18.25 (4.75) 18.69 (4.38) Physical Neglect 12.18 (1.82) 11.90 (1.83) 12.11 (1.83) Emotional Abuse 8.83 (3.85) 9.96 (4.77) 9.11 (4.19) Physical Abuse 6.25 (2.65) 6.28 (2.71) 6.26 (2.66) Sexual Abuse 10.30 (3.04) 9.75 (2.86) 10.17 (3.00)

(26)

Table 2.

Items of the Positive and Negative Syndrome Scale (PANSS) and the Childhood Trauma Questionnaire - Short Form (CTQ-SF) and their assigned colors and labels

Item label Domain color Item description

P1 Purple Delusions

P2 Purple Conceptual Disorganization

P3 Purple Hallucinations

P4 Purple Hyperactivity

P5 Purple Grandiosity

P6 Purple Paranoia

P7 Purple Hostility

N1 Blue Blunted Affect

N2 Blue Emotional Withdrawal

N3 Blue Poor Rapport

N4 Blue Social Withdrawal

N5 Blue Difficulty in Abstract Thinking

N6 Blue Lack of Spontaneity and Flow of Conversation

N7 Blue Stereotyped Thinking

GP1 Yellow Somatic Concern

GP2 Yellow Anxiety

GP3 Yellow Guilt

GP4 Yellow Tension

GP5 Yellow Mannerism and Posturing

GP6 Yellow Depression

GP7 Yellow Motor Retardation

GP8 Yellow Uncooperativeness

GP9 Yellow Unusual Thought Content

GP10 Yellow Disorientation

GP11 Yellow Poor Attention

GP12 Yellow Poor Judgment and Insight GP13 Yellow Disturbed Willpower GP14 Yellow Poor Impulse Control

(27)

GP16 Yellow Active Social Avoidance

CT1 Maroon Physical Neglect

CT2 Maroon Emotional Neglect

CT3 Maroon Sexual Abuse

CT4 Maroon Emotional Abuse

(28)

Supplementary Materials

Supplementary Figure 1. Centrality measures of the network including two scales of the

(29)

Supplementary Figure 2. Centrality measures of the network including the three scales of

the Positive and Negative Syndrome Scale (PANSS): positive symptoms, negative symptoms, and general psychopathology

(30)

Shortest Paths: Emotional Abuse – Positive and Negative Symptoms

Supplementary Figure 3. Network depicting shortest paths between the Emotional Abuse

scale (i.e., node CT4) of the Childhood Trauma Questionnaire - Short Form (CTQ-SF), and the two main dimension of the Positive and Negative Syndrome Scale (PANSS): positive symptoms and negative symptoms (i.e., the nodes through which one has to pass to reach all positive and negative symptoms from the sexual abuse dimension). Dashed and dotted lines represent background connections existent within the network, that are not relevant when investigating shortest paths; bigger dots represent stronger connections, while smaller dots and dashed lines represent weaker connections.

(31)

Shortest Paths: Emotional Neglect– Positive and Negative Symptoms

Supplementary Figure 4. Network depicting shortest paths between the Emotional Neglect

(32)

Shortest Paths: Physical Neglect – Positive and Negative Symptoms

Supplementary Figure 5. Network depicting shortest paths between the Physical Neglect