Routes to psychotic symptoms: Trauma, anxiety and psychosis-like experiences

1Psychosis is likely to arise from a number of interacting factors. One potential factor that has gained some prominence recently is trauma. A number of studies, in both clinical and non-clinical populations, have found an association of trauma and psychotic symptoms (e.g. ). For instance,, in a longitudinal survey of 4000 adults in the general population, found that a history of childhood abuse increased the risk for the subsequent development of psychotic symptoms by over 10 times. This has led to theoretical speculation about how trauma may increase the likelihood of delusions and hallucinations (e.g. ) and a smaller number of empirical studies investigating the ideas (e.g. ).

2The evidence for the association between trauma and psychosis is strongest for the occurrence of hallucinations. If trauma has a causal role, then its influence on delusions and hallucinations may be somewhat different. For instance, the role of trauma in hallucinations may be quite direct, whereby trauma-related intrusions and flashbacks enter the content of hallucinations (e.g. ). On the other hand the influence of trauma on paranoia may be non-specific. The experience of trauma is likely to create negative ideas about the self, anxiety, and depression, which are known risk factors for paranoia (e.g. ). Evidence consistent with these hypotheses was recently found by our research group in a cross-sectional study of 200 students.

3In this article, we report a replication study, but with a number of significant improvements: the sample studied was broadly representative of the general population for socio-economic status; the intellectual functioning of participants was known; new measures that capture clear persecutory and hallucinatory experiences were used; traumatic events that met a diagnostic criterion for severity were assessed; and there were additional measures of depression and anxiety. It was predicted that previous experience of severe traumatic events would be associated with current levels of non-clinical paranoia and hallucinations. Further, it was predicted that experience of trauma would have its effect on paranoia via affect, specifically anxiety and negative beliefs about the self. Finally, a related route by which trauma may influence psychosis was assessed: use of illicit drugs. Both trauma and psychosis are associated with the use of substances such as cannabis (e.g. ). It is plausible that coping with the emotional consequences of trauma by taking illicit drugs could contribute to the development of psychotic-like experiences. However, causal claims from a cross-sectional study are clearly limited.

4Two hundred members of the general public completed a trauma assessment and symptom measures as part of a virtual reality study of paranoia. The current report concerns unreported tests of associations between the baseline assessments used in the study.

5A representative sample of the UK adult local population was recruited via distribution of a leaflet to 20,000 households in local south London postcodes. The postcodes score highly on indexes of deprivation, consistent with the inner city location. Approximately 350 people responded to the advertisement. The leaflets concerned ‘virtual reality research at King's College London' and did not mention paranoia, hallucinations, trauma or psychiatry. Seven individuals reporting a history of Axis-1 severe mental illness (e.g. schizophrenia, bipolar disorder) were excluded. Two individuals with a history of epilepsy were also excluded because of potential virtual reality side effects. One hundred male and one hundred female participants were recruited. The occupationally based National Statistics Socio-economic Classification was used to categorise participants. The study had received approval from the research ethics committee of King's College London.

6The checklist asks respondents about the occurrence of a range of severe life events (e.g. serious accident, physical attack, sexual abuse). If the respondent reports the occurrence of an event, subsequent questions ask when the event happened, whether the person thought at the time serious harm or death could result, and whether feelings of intense helplessness, fear or horror occurred. Only checklist items that referred to Criterion A events as defined in the Diagnostic and Statistical Manual of Mental Disorders were used. Only events that reached the severity criterion related to post-traumatic stress disorder diagnosis were scored as occurring. The total number of traumatic events, the total number of victimisation events, the number of childhood traumatic events, and the number of traumatic events in the past year were recorded. report adequate test–retest reliability of the measure in a sample of 200 women.

7The G-PTS Part B is a trait measure of persecutory ideation. Each of the 16 items (e.g. ‘I was convinced there was a conspiracy against me’, ‘Certain individuals have had it in for me’, ‘I have definitely been persecuted’) conforms to a recent definition of persecutory ideation and is rated on a scale from 1 to 5. The presence of persecutory ideation is assessed over the past month and higher scores indicate greater levels of persecutory thinking. The questionnaire has been psychometrically evaluated for use in both clinical and non-clinical populations. The internal consistency of the scale and the test–retest reliability are good. Convergent validity with the Paranoia Scale has been shown.

8This 32-item questionnaire, developed in both non-clinical and psychosis groups, assesses a range of perceptual anomalies. The measure shows good construct and criterion validity, internal reliability and test–retest reliability. In the current study we focussed upon the three verbal hallucinations items together (‘Do you ever hear voices commenting on what you are thinking or doing?’, ‘Do you ever hear voices saying words or sentences when there is no one around that might account for it?’, ‘Have you ever heard two or more unexplained voices talking with each other?’). The scale also has three factor scores. The first factor, temporal lobe experience, contains items such as ‘Do you ever think that everyday things look abnormal to you?’ and ‘Do you ever see shapes, lights, or colors even though there is nothing really there?’. The second factor, chemosensation, contains items such as ‘Do you ever notice that food or drink seems to have an unusual taste?’ and ‘Do you ever smell everyday odors and think that they are unusually strong?. The third factor, clinical psychosis, contains items such as ‘Do you ever hear your own thoughts spoken aloud in your head, so that someone near might be able to hear them?’ and ‘Do you ever hear voices commenting on what you are thinking or doing?’. Respondents initially rate (No/Yes) whether they have experienced the anomaly. A higher score represents the reporting of a greater number of perceptual anomalies.

9The WASI is a nationally standardised short and reliable measure of intelligence linked to the Wechsler Adult Intelligence Scale—Third Edition. The Vocabulary and Matrix Reasoning subtests were used in the current study.

10The DASS is a 42-item instrument with three subscales measuring current symptoms of depression, anxiety, and stress. Each of the subscales consists of 14 items with a 0–3 scale (0 = did not apply to me at all, 3 = applied to me very much). Higher scores indicate higher levels of emotional distress. The scale has been shown to be reliable and valid in a large UK non-clinical population. The anxiety and depression subscales were used in the current study.

11This measure, developed with non-clinical and psychosis groups, has 24 items each rated on a 5-point scale (0–4). Four subscale scores are derived: negative beliefs about self, positive beliefs about self, negative beliefs about others and positive beliefs about others. Higher scores reflect greater endorsement of items. In this study the scale Negative beliefs about self was of interest.

12The MAP was developed with a large sample from a substance abuse clinic. Respondents are asked directly about the use over the past month of illicit drugs including cannabis, cocaine powder, crack cocaine, heroin, amphetamines, and methadone. Use of illicit drugs was dichotomised into absence and occurrence (meaning that any level of drug use in the past month would count as occurrence).

13All analyses were carried out using. All hypothesis testing was two-tailed, and 95% confidence intervals (CI) are reported. The key variables of history of trauma, paranoid ideation, and verbal hallucinations were dichotomised (presence/absence) and relationships between these variables were then assessed with Chi-square tests. (As a check that the associations were not simply determined by demographic differences, binary logistic regressions controlling for age, sex, ethnicity, education level, socio-economic status, and intellectual functioning were then carried out.) The variables anxiety, depression, negative ideas about the self, and illicit drug use were then used to understand the nature of any significant associations between trauma and psychotic-like experiences. The putative mediation variables were first tested for associations with trauma and symptoms using t-tests or Chi-square. Then, two binary logistic regressions were carried out with paranoia and hallucinations as the dependent variables, and lifetime trauma and potential mediator variables as the independent variables. These regressions were also repeated controlling for demographic information.

14The average age of the participants was 37.5 (S.D. = 13.3) (minimum = 18, maximum = 77). The mean IQ score was 104.6 (S.D. = 12.0) (minimum = 69, maximum = 133). Further basic information on the participants is presented in Table 1. There is a spread of participants across socio-economic categories, and the proportion in each category is broadly representative of the United Kingdom population.

15At least one traumatic event had been experienced by 70% of the sample (n = 140). Childhood physical or sexual abuse had been experienced by 25.5% (n = 51), with 7.5% experiencing severe childhood sexual abuse (n = 15). A traumatic event had been experienced in the last year by 15%. Endorsement rates for individual events are presented in Table 2.

16The mean persecutory ideation score was 22.6 (S.D. = 11.6) (range 16–78). Paranoid ideation was not reported by 42.5% of the sample(n = 85), and the scores showed an exponential distribution similar to other studies. Verbal hallucinations were reported by 15.5% of the sample (n = 31), and at least one anomalous experience in the clinical psychosis factor was reported by 44.5%r (n = 89). There were higher rates for endorsement of at least one item for the temporal lobe (75%) and chemosensation factors (72%).

17The occurrence of at least one lifetime traumatic event was associated with a 2.5 times greater risk of endorsing a persecutory thought than not reporting any traumatic event (see Table 3). Similarly, the occurrence of at least one lifetime traumatic event was associated with a 4.8 times greater risk of endorsing a verbal hallucinations item. (These associations of lifetime trauma and psychosis-like experiences remained significant in logistic regressions controlling for age, sex, ethnicity, education level, socio-economic status, and intellectual functioning.) It is also of note that a history of lifetime trauma was associated with higher scores on the Cardiff Anomalous Perception Scale factors of temporal lobe experience, chemosensation, and clinical psychosis (all P-values < 0.01).

18Type of trauma was also examined (Table 3). Using the whole sample, it can be seen that non-victimisation events and severe childhood sexual abuse were particularly associated with psychotic-like experiences. Recent adult traumas did not have an impact. When the group with no history of any trauma was used as the comparison (n = 60), all trauma types, except adult history of physical attack, were significantly associated with the presence of paranoid ideation and verbal hallucinations (P < 0.05). The analysis was also repeated removing the individuals without a history of trauma (i.e. looking within the group who had experienced at least one trauma). Experience of non-victimisation events was significantly associated with both paranoid ideation and verbal hallucinations (P < 0.05). Severe childhood abuse was strongly associated with verbal hallucinations (P = 0.001), and almost had a near-significant association with paranoia (P = 0.056). All other comparisons were non-significant (P > 0.05).

19The presence of at least one lifetime trauma was significantly associated with higher levels of anxiety, depression and negative ideas about the self (all P-values < 0.05) (see Table 4). Lifetime trauma was also associated with current use of illicit drugs, Chi-square (df = 1) = 4.586, P = 0.032, odds ratio = 2.14, CI = 1.06, 4.32. The presence of paranoid ideation was significantly associated with higher levels of anxiety, depression and negative ideas about the self (all P-values < 0.01). Persecutory ideation was also associated with current use of illicit drugs, Chi-square (df = 1) = 8.640, P = 0.003, odds ratio = 2.57, CI = 1.36, 4.88. The presence of verbal hallucinations was significantly associated with higher levels of anxiety (P < 0.05), had a trend to be associated with higher levels of depression (P = 0.068) but was not associated with negative beliefs about the self (P = 0.122). Verbal hallucinations were not significantly associated with current use of illicit drugs, Chi-square (df = 1) = 0.749, P = 0.387, odds ratio = 0.68, CI = 0.29, 1.62. Illicit drug use was associated with higher levels of anxiety (P = 0.006), but not significantly with levels of depression (P = 0.058) or negative ideas about the self (P = 0.237).

20Only potential mediation variables that had a significant association with trauma and paranoia or hallucinations were entered into the next stage of the analysis. A binary logistic regression was carried out with paranoid ideation as the dependent variable and a history of trauma, anxiety, depression, negative ideas about the self, and illicit drug use entered as independent variables (see Table 5). Only anxiety was a significant predictor in this model. A similar finding was obtained including age, sex, ethnicity, education, intellectual functioning, and socio-economic status as covariates. A further binary logistic regression was carried out with verbal hallucinations as the dependent variable and a history of trauma and anxiety as independent variables. Both lifetime trauma and anxiety were significant predictors (see Table 5). When age, sex, ethnicity, education, intellectual functioning, and socio-economic status were entered as covariates, then lifetime trauma remained significant (P = 0.021), but the association with anxiety became a trend (P = 0.055).

21The study provides further confirmation that lifetime experience of trauma is common in the general population. Seventy percent of the general population sample had experienced a trauma, although this rate is much higher than the findings from large epidemiological studies, and severe traumas such as childhood sexual abuse were present in a much smaller proportion of the group. However the focus of the current study was not on trauma prevalence but on identifying links between trauma and instances of psychotic-like phenomena. Clear associations were found. Experiencing a trauma was associated with almost a five times greater likelihood of verbal hallucinations and approximately half that risk of paranoid thoughts. These results are consistent with previous research. It is difficult to disentangle the effect of different trauma types, but in this sample there was evidence that psychotic-like experiences had the strongest associations with severe childhood sexual abuse and, unexpectedly, non-victimisation events.

22Trauma may impact in different ways on persecutory thinking and hallucinatory experience. Previous research has provided strong evidence of a link between anxiety and persecutory thinking. Anxiety has repeatedly been found to be associated with paranoid thoughts (e.g. ) and persecutory delusions (e.g. ). Further, anxiety is predictive of the occurrence of paranoid thoughts (e.g. ) and of the persistence of persecutory delusions. Moreover, it has been shown in non-clinical groups that paranoid thoughts build upon common interpersonal anxieties and worries. The results of the current study indicate that trauma influences persecutory thinking non-specifically via the creation of anxiety. Anxiety also accounted for the association of substance use and paranoia. Anxiety may be a common linking factor between trauma, illicit drug use, and paranoia. However it is entirely plausible, and supported by the presence of a statistical trend in the mediational analysis, that the role of illicit drug use in paranoia has a degree of independence from anxiety.

23In contrast the link between trauma and hallucinations was unexplained by the small number of putative mediational variables assessed in the current study. A history of trauma and levels of anxiety were independently associated with the occurrence of verbal auditory hallucinations. In this study we clearly did not succeed in including variables relevant to the association between trauma and hallucinations. Specific assessments of trauma-related intrusions and flashbacks were needed to have tested hypothesised direct links between trauma and hallucinations (e.g. ), although it might be argued that levels of anxiety are a proxy for, or at least highly correlated, with these processes. It would also have been of interest to have included assessments relevant to inner speech and self-monitoring models of voices (e.g. ).

24There are weaknesses in the study that merit caution in interpreting the results. Foremost, the cross-sectional study design makes both causal and mediation claims weak. There was multiple statistical testing raising the likelihood of the occurrence of Type I errors. It is also unlikely that the sample is truly representative of the UK general population. The reliance on self-report is important to note. This is most problematic for persecutory thinking. It is plausible that there was overlap between responses for the paranoia and trauma questionnaires. That is, a proportion of the persecutory thinking may not have been unfounded and instead related directly to the past trauma. In the experimental study the participants took part in, where unfounded paranoia was assessed, no direct association of trauma and paranoia was identified. This highlights the need for the study of trauma and psychosis to go beyond the establishment of cross-sectional self-report associations.