Early trauma and schizophrenia onset: preliminary results of an outpatient cohort in Brazil

Abstract Objectives To assess the prevalence of early trauma in individuals with onset of schizophrenia (SZ) at early (≤ 18 years) and adult (> 18 years) ages (EOP and AOP, respectively) and explore relationships between the onset of disease and clinical variables including traumatic events and psychotic and mood symptoms. Methods Subjects with SZ (n = 71) and EOP and AOP were compared for history of psychological trauma, sexual abuse, and physical punishment using the Early Trauma Inventory Self Report - Short Form (ETISR-SF). They were also compared for history of comorbidities and affective disorders using the Diagnostic Interview for Psychosis and Affective Disorders, the Positive and Negative Syndrome Scale, the Liebowitz Social Anxiety Scale, and the Calgary Depression Scale for Schizophrenia. Coefficients were calculated for correlations between scale results and disease duration. Results Early trauma was significantly associated with an early onset psychotic episode (r = -0.315, p < 0.01). General trauma and depressive symptoms in adulthood were also associated (r = 0.442, p < 0.01), as were social anxiety symptoms and early trauma (r = 0.319, p < 0.01). Total ETISR-SF scores and the physical abuse item were significantly higher in EOP than in AOP. In the hierarchical regression, PANSS scores were best predicted by a model including the duration of disease and age of first psychotic episode (R = 0.303). Conclusions Our results support the hypothesis that early trauma, including physical abuse, may play a relevant role in schizophrenia symptoms, such as an earlier psychotic occurrence, as well as features of other psychiatric disorders, such as greater severity of social anxiety and depression.


Introduction
Schizophrenia (SZ) is a severe psychiatric condition characterized by psychosis, a decline in function and deficits in social and emotional function, and cognitive impairments. 1 The worldwide prevalence of schizophrenia is estimated at 0.28 to 0.7%, with onset most typically occurring in late adolescence and mid-30s. 2 Although pathogenic mechanisms are unknown, an interaction between genetic, molecular, environmental, and neurodevelopmental factors seems to be relevant. [1][2][3][4] Advances in genetic research have shown that the most influential association with schizophrenia at a population level involves variants of the Major Histocompatibility Complex (MHC) locus, more Alves specifically, in parts of various structurally diverse alleles of complement component 4 (c4) genes. 4 In this context, environmental exposure to psychological stress factors with early trauma can also increase the risk of phenotypic expression of the disease. 5 Emotionally traumatic experiences occurring in early life stages seem to influence the clinical features of schizophrenia, apparently promoting greater severity of cognitive deficits and productive symptoms, including hallucinations and delusions. 6  of the HPA axis. 7 Prenatal stress, maternal separation, and childhood abuse and emotional neglect are associated with the severity of both psychotic and depressive symptoms in adulthood and an overall reduction in brain volumes and increased amygdala compared to the whole brain volume. 8 Mood symptoms and anxiety also seem to be strongly influenced by traumatic experiences, with correlations with premorbid symptoms and comorbidities in schizophrenia.
Depressed mood, anhedonia, and social withdrawal, which are frequent in schizophrenia, 9,10 could be related to early trauma, as observed in post-traumatic stress disorder (PTSD). 11 Moreover, anxiety and depression might be influenced by the prevalence of early trauma in SZ, and may be related to neurobiological changes, such as HPA axis hyperactivity. 11 Early life experiences influence the development of relationship patterns in adulthood and might also do so among those with schizophrenia. 10 Another important topic is how early exposure to traumatic experience may influence the onset of SZ symptoms. There are several hypotheses associating premorbid factors with early onset of psychosis.
Epidemiological studies have suggested that psychoticlike experiences are more common by the age of 12 in children previously exposed to childhood trauma. 12  Furthermore, EOP patients were likely to have worse premorbid functioning and a significantly longer duration of untreated psychosis than those with later onset. 18,19 The overarching aim of this study was to assess the prevalence of early trauma in individuals with schizophrenia and examine the relationship between these events and symptoms in schizophrenia, including psychiatric comorbidities, and to test the hypothesis of whether early exposure to psychological or physical abuse was associated with younger age of onset and more severe clinical features.

Method
The  were deemed low, 28 and subjects tended to exhibit low to middle scores on the Liebowitz scale for fear/anxiety and avoidance. 24 On the PANSS scale, negative symptoms, such as emotional withdrawal and lack of attention, were more frequent than positive symptoms ( Table 2). Table 3 shows the group comparisons for disease onset. AOP individuals were significantly older than EOP participants, although they had presented the disease for fewer years. Total ETISR-SF scores and the physical abuse item were significantly higher among EOP than AOP. However, there were no statistical differences between the groups in Calgary or Liebowitz scale scores.

ETISR-SF and age at disease onset analysis
For the entire sample (Table 4), the ETISR-SF sexual abuse score was positively correlated with the duration (in years) of the disease (r = 0.327, p ≤ 0.01);   ETISR-SF emotional abuse also correlated inversely with age at first psychotic episode (r = -0.249, p ≤ 0.05). Table 5 shows correlations between ETISR-SF scale results and the clinical characteristics of groups.
More frequent report of sexual abuse was correlated with longer disease duration in the EOP and with earlier breakdown of psychosis in the AOP; more frequent report of physical abuse was associated with longer disease duration in the AOP.

ETISR-SF and Calgary, Liebowitz and PANSS analysis for the entire sample
Total PANSS score was inversely correlated with age

ETISR-SF and Calgary, Liebowitz, and PANSS analysis for EOP and AOP subsets
Correlations for the EOP and AOP groups can be found in Table 4. The item fear and anxiety from the Liebowitz scale shows a positive correlation with ETISR-SF reaction to trauma in the EOP group (Table 4).
Positive symptoms on the PANSS were correlated with younger disease onset in both the EOP (r = -0.364; p < 0.05) and AOP groups (r = -0.365; p < 0.05).
Conversely, negative symptoms on the PANSS were negatively correlated with disease duration in the EOP group (r = -0.499; p < 0.01) and positively with age in the AOP group (r = 0.351; p < 0.05). No significant correlations were found between positive or negative PANSS symptoms and the ETISR-SF subdomains in either the EOP or the AOP groups.

Hierarchical regression models
In the hierarchical regression, only two models were identified as potential predictors of PANSS scores (Table   6). Of these, the model that achieved better prediction for the PANSS score was the one combining duration of disease and age of first psychotic episode, although with only 6.3% of the variance (Table 6).

Discussion
The current study investigated the prevalence of early trauma in schizophrenia and its relationships with clinical symptoms and with the age of onset of the disease and highlighted significant findings: first, there was an association between higher levels of trauma in the ETISR-SF, particularly the emotional abuse dimension, and earlier onset of schizophrenia; also, history of sexual abuse was associated with longer duration of disease in the EOP and with earlier onset of the disease in the AOP; finally, the prevalence of positive symptoms was associated with younger age in both early and adult-onset groups (Figure 1).  Diagnosis of depression could be more closely related to severe physical abuse than to less severe maltreatment forms. 16,32 Also, early exposure to trauma and stress predicted higher depression scores, also measured by the dysthymia category on the PANSS scale. 8

Positive-negative symptoms and traumata
Regarding the prevalence of positive-negative