25 June 2005. Does exposure to psychological trauma in childhood—war, natural disasters, child abuse—cause schizophrenia? A new study that assessed childhood trauma, then prospectively measured the appearance of psychotic symptoms, doesn’t answer that question directly, but it suggests that for at least a subset of schizophrenia symptoms, the answer is yes. The work, from Jim van Os at the University of Maastricht, the Netherlands, with Roselind Lieb and Hans-Ulrich Wittchen at Max Planck Institute, Munich, showed a dose-response relationship between the severity of self-reported childhood trauma and later psychoses in a large group of young Germans. The effect of trauma was enhanced in subjects who measured high on a scale of psychotic proneness at the beginning of the study. The results, appearing in the June issue of the British Journal of Psychiatry, support the idea that childhood trauma can promote psychosis later in life, and in particular may bring out or exacerbate symptoms in a subset of susceptible people.
While psychological trauma is clearly associated with depression and other symptoms of post-traumatic stress disorder, a link to psychosis has been unclear. Many previous studies have surveyed populations with existing mental illness and found a high incidence of a history of childhood physical or sexual abuse. To prospectively look for an association between self-reported traumatic experiences and the risk of psychosis, first author Janneke Spauwen and colleagues interviewed 3,021 subjects between 14 and 24 years old at entry to determine their experience of traumatic events. Showing a list of nine horrors ranging from war experience, abuse, rape, and natural catastrophe to a serious accident, the interviewers elicited histories from the subjects. The presence of psychosis or a proneness to psychosis was assessed with a checklist of psychosocial effects, including scales for hallucinations, delusions, paranoia, and psychoticism. Subjects were reassessed an average of 3.5 years later.
The data, based on follow-up of 2,524 men and women, showed that exposure to any trauma at all increased the risk of psychosis, and the effect was strongest for the most stringent definition of psychosis (three or more psychotic symptoms reported vs. zero, one, or two). The odds ratio (adjusted for confounding factors and psychosis proneness at the beginning of the study) was 1.89 (95 percent confidence interval 1.16-3.08) for exposure to any trauma and the occurrence of at least three psychotic symptoms. Beyond the broad categories of trauma, the study did not measure the type of trauma in great detail, except to distinguish severe trauma—engendering feelings of intense fear, helplessness, or horror—which was more strongly associated with later psychosis. The association was the same whether the trauma occurred before age 13 or later in life.
A dose response effect supported the idea that trauma played a causal role in the psychotic symptoms. The odds ratio for psychosis increased with an increasing number of trauma events. The effect of trauma appeared specific for psychosis, as the researchers found no association between self-reported trauma and the occurrence of bipolar disorder or major depression.
Experiencing traumatic events was significantly more likely to produce psychoses in people who scored in the top quartile on tests of psychosis proneness, compared to those in the lower three quarters. The effect size for trauma in those without psychosis proneness was 1.8 percent, while in those with proneness, it was 7 percent. The difference was statistically significant, and showed a synergistic rather than additive effect of childhood trauma on later psychoses in this susceptible group. Of course, the authors concede, childhood abuse itself could affect psychosis proneness, and they did detect a weak interaction between reported trauma at baseline and signs of psychosis proneness.
The results are entirely consistent with many previous retrospective studies showing high levels of childhood abuse and trauma among schizophrenia patients. By relaxing the outcome criteria from a diagnosis of schizophrenia to the presence of any of a broad category of symptoms, the researchers were able in this study to prospectively and convincingly link trauma to a later psychotic state. A handful of other studies have recently used similar designs and along with this report, the results are starting to show a consistent pattern between trauma and psychosis, particularly hallucinations (reviewed in Read et al., 2005). It remains to be seen, however, if the association between trauma and psychotic symptoms also applies to psychotic disorders, including schizophrenia.
More work will be necessary to sort out the mechanisms by which trauma increases the risk of psychosis, but several likely hypotheses exist. Childhood stress affects early brain development, and changes the sensitivity of the hypothalamic-pituitary-adrenal (HPA) axis. This could contribute to the dopamine abnormalities seen in psychosis. It is also possible that traumatic experiences early in life could directly affect dopamine pathways.
In any case, the study reveals childhood trauma as an often-hidden factor that could explain much of the burden of psychosis that occurs outside of major mental disorders. But what about the psychosis of schizophrenia? As van Os and colleagues have written (Read et al., 2005), the practice of classifying psychotic symptoms as PTSD whenever there is clear evidence of trauma may have led clinicians astray and hindered studies on the possible role of trauma in raising the risk of schizophrenia-like disorders. The current work bolsters the case for taking a symptom-centric view of stress-related mental illnesses, especially in light of the fact that different diagnoses, including post-traumatic stress disorder and schizophrenia, involve common psychotic hallmarks like hallucinations and paranoia.—Pat McCaffrey.
Spauwen J, Krabbendam L, Lieb R, Wittchen HU, van Os J. Impact of psychological trauma on the development of psychotic symptoms: relationship with psychosis proneness. Br J Psychiatry. 2006 Jun;188:527-33. Abstract