Letter to the Editor: Childhood trauma as a risk factor for psychosis: the confounding role of cognitive functioning

Introduction

Van Dam et al. (2015) describe childhood trauma as a contributor to vulnerability for psychosis and depressive symptoms. In this context, the role of cognitive functioning deserves attention, as cognitive impairment precedes the onset of psychosis (Bora et al. 2014; Agnew-Blais et al. 2015) and is associated with childhood trauma (Aas et al. 2012, 2014). The neurodevelopmental view of psychosis states that negative environmental influences, such as trauma, in addition to early developmental (genetic) abnormalities give rise to psychotic symptoms (Read et al. 2005; Owen et al. 2011).

The underlying mechanisms of the association between childhood trauma and increased risk for psychosis are unknown. Early traumatic events are associated with reduced cognitive functioning in the domains of executive functioning, working memory and (verbal) intellectual abilities, which is similar to cognitive profiles of patients with psychotic disorders and ultra-high-risk individuals (Reichenberg, 2005; Aas et al. 2012). It can therefore be hypothesized that childhood trauma exerts neurobiological consequences that contribute to psychosis-associated cognitive impairment.

Our group studied psychotic experiences such as auditory verbal hallucinations in non-clinical individuals (Sommer et al. 2010), without a history of hospitalization, anti-psychotic medication use, or negative symptomatology. This population shows reduced performance in the cognitive domains of executive functioning, working memory and language (Daalman et al. 2011) and, interestingly, reports high rates of childhood trauma, similar to patients with a psychotic disorder (Daalman et al. 2012). We investigated whether increased exposure to early traumas can explain the reduced cognitive performance associated with a vulnerability to experience psychotic symptoms, in non-clinical adults with and without psychotic experiences.

Method

Participants and procedures

We included 101 non-clinical individuals with psychotic experiences (auditory verbal hallucinations) and 101 controls (selection procedures and inclusion criteria described previously; Daalman et al. 2011, 2012). Participants did not meet criteria for Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) diagnosis, yet depression in remission was allowed. Neuropsychological assessment included several cognitive domains (Daalman et al. 2011). Childhood trauma was rated using the Childhood Trauma Questionnaire Short Form (CTQ-SF; Bernstein et al. 2003). Both groups were similar with regard to gender, age, handedness and education (Daalman et al. 2011).

Statistics

Correlations between the CTQ-SF and cognitive test scores were calculated using Spearman's rho (ρ).

To address the effect of childhood trauma, we conducted step-wise regression analyses for those cognitive measures that were significantly lowered in individuals with psychotic experiences in our previous study (Daalman et al. 2011). Group was included as a predictor in the first step and CTQ-SF score was added in the second step, for verbal inhibition (Stroop interference), working memory [digit-span backward; Wechsler Adult Intelligence Scale-III (WAIS-III)], verbal abilities (vocabulary and similarities; WAIS-III) and verbal intelligence (National Adult Reading Task). If trauma severity is a significant factor in explaining the association between reduced cognitive performance and psychotic experiences, adding CTQ-SF score to the regression model will reduce the significance level of this association (partial explanatory variable) or render it non-significant (full explanatory variable).

Ethical standards

All procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.

Results

Individuals with non-clinical psychotic symptoms reported more childhood trauma (CTQ-SF: mean = 45.1, s.d. = 15.7), compared with controls (mean = 36.4, s.d. = 8.6, t = 23.7, p < 0.001).

Associations between cognitive performance and CTQ-SF score were corrected for multiple analyses, using Bonferroni correction. Given the 14 cognitive tests, significance level was set at p < 0.0036. Childhood trauma was related to reduced verbal inhibition (Stroop interference: ρ = 0.210, p = 0.003). Childhood trauma was unrelated to other cognitive measures (ρ = −0.167–0.110, all p > 0.0036). Childhood trauma fully explained the previously observed association between reduced verbal inhibition and non-clinical psychotic features (Daalman et al. 2011), as group was no longer a significant predictor for Stroop interference when adding CTQ-SF score (Table 1). Childhood trauma also accounted for the association between non-clinical psychotic symptoms and reduced working memory. Childhood trauma was of no explanatory value with regard to the association between psychotic features and reduced verbal abilities, nor intelligence (Table 1).

Table 1. Step-wise regression analyses evaluating the effect of childhood trauma on the reduced cognitive functioning associated with non-clinical psychotic experiences

s.e., Standard error; CTQ-SF, Childhood Trauma Questionnaire Short Form; WAIS, Wechsler Adult Intelligence Scale.

^a Given the five cognitive tests, we corrected for multiple analyses using Bonferroni correction by dividing the significance level of 0.05 by 5, therefore set at p  < 0.01.

Discussion

In adults with and without non-clinical psychotic experiences, we observed a correlation between level of childhood trauma and reduced verbal inhibition in the domain of executive functioning. Childhood trauma fully explained the previously observed associations between psychotic experiences and reduced executive functioning, as well as lower working memory performance. Childhood trauma was not relevant for the relationships between non-clinical psychotic experiences and reduced verbal abilities or intelligence.

Our results extend the finding of van Dam et al. (2015) that childhood trauma increases the vulnerability for psychotic symptoms later in life, by showing its potential impact on executive functioning and working memory performance. These cognitive domains are largely supported by the prefrontal cortex, just like many symptoms of primary psychotic disorder resemble those associated with frontal abnormalities, including reduced spontaneity, mental rigidity and lack of social judgment (Reichenberg, 2010). The prefrontal lobes play a central role in the neurodevelopmental hypothesis of psychosis, as they are especially vulnerable to stress and mature late, in the phase when brain abnormalities following genetic and environmental influences are hypothesized to interfere with normal brain development (Reichenberg, 2010).

Interestingly, van Dam et al. (2015) also show a vulnerability after trauma for depressive symptoms. Several psychiatric disorders are characterized by impairments in executive functioning and working memory as well as increased exposure to trauma, such as depression, post-traumatic stress disorder and bipolar disorder (Gould et al. 2012). In these conditions, the mechanisms may be similar, which implies that prevention of childhood trauma could be a means to forestall severe dysfunction from cognitive impairments in psychiatric disorders.

Our results indicate that reduced cognitive functioning, especially executive functioning and working memory, may be part of the causal pathway from childhood trauma to psychotic symptoms later in life. We emphasize the need for longitudinal studies to establish whether neurobiological consequences of early trauma increase the vulnerability for psychosis via a negative impact on cognition.