A history of trauma, particularly childhood adversity, among patients is a recognised risk factor for developing psychosis.Reference Varese, Smeets, Drukker, Lieverse, Lataster and Viechtbauer1
Although evidence suggests that childhood trauma is associated with brain alterations in psychosis,Reference Cancel, Dallel, Zine, El-Hage and Fakra2 the association between psychotic symptoms and trauma history is not clearly understood at both clinical and pathophysiological levels.
A history of trauma has been preferentially associated with positive psychotic symptoms, notably in patients with first-episode psychosis (FEP).Reference Ajnakina, Trotta, Oakley-Hannibal, Forti, Stilo and Kolliakou3 However, the links between specific types of positive psychotic symptoms and specific types of traumas remain unclear. Some studies found associations between them, in particular hallucinations and/or delusions.Reference Read and Argyle4 Bentall and colleagues found associations between a history of childhood rape and auditory-verbal hallucinations, between institutionalisation and paranoid ideation, and between childhood physical abuse and auditory-verbal hallucinations and paranoid ideation.Reference Bentall, Wickham, Shevlin and Varese5 In other studies, no relationship between particular types of trauma and psychotic symptoms was found.Reference Van Nierop, Lataster, Smeets, Gunther, Van Zelst and De Graaf6 Some authors reported on the impact of multiple trauma exposure on psychosis rather than single events.Reference Longden, Sampson and Read7
Few qualitative studies were conducted in the field. They found thematic associations between trauma history and positive symptoms,Reference Peach, Alvarez-Jimenez, Cropper, Sun, Halpin and O’Connell8 trauma and auditory hallucinations,Reference Hardy, Fowler, Freeman, Smith, Steel and Evans9 and trauma and hallucinations.Reference Rosen, Jones, Longden, Chase, Shattell and Melbourne10
Few qualitative studies have investigated patients’ views on the association between positive psychotic symptoms and traumatic history.Reference Van Sambeek, Franssen, Van Geelen and Scheepers11–Reference Campodonico, Varese and Berry13 To our knowledge, none were specifically conducted in patients with FEP.
The current study aimed to describe the experience of positive psychotic symptoms and its association with different types of life and trauma events, considering the timing of exposure from the perspective of patients who have experienced FEP.
Method
Semi-structured interviews were conducted to gather the patients’ perspective on positive symptoms and links to their life and trauma history. Qualitative studies are particularly relevant when a topic is still relatively unexplored. They enable us to generate hypotheses that can then further be tested quantitatively. They provide access to detailed individual experience in order to understand a phenomenon.
Setting and sample
Participants were recruited via the ‘PEP’S’ FEP programme in Greater Lyon (France). Inclusion criteria were as follows: to have experienced FEP with a diagnosis made by an FEP-specialised psychiatrist, including those induced by post-traumatic stress disorder and those with or without affective symptoms; to speak and read French; to be aged between 18 and 35 years and to consent to the study. Exclusion criteria were: having an intellectual disability (<70 IQ on premorbid intelligence estimate), patient refusal, adults under guardianship or tutorship, disorganisation greater than 4 on the Comprehensive Assessment of At-Risk Mental States (CAARMS),Reference Krebs, Magaud, Willard, Elkhazen, Chauchot and Gut14 an ongoing suicidal crisis, and hearing or vision problems that could compromise the interview process. All participants meeting the inclusion criteria and involved in the PEP’S programme were considered for inclusion. All eligible participants were contacted by their case manager during their appointments or shortly before discharge from hospital. Recruitment took place on an ongoing basis until the appropriate sample was reached. Analysis was performed continuously, and recruitment was stopped when data saturation was reached.
Ethical considerations
The authors assert that 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 2013. All procedures involving human patients were approved by an Institutional review Board (Comité de Protection des Personnes Ile de France VII, approval number 2022-A01780-43). All participants were informed of the qualitative aspects of the interview, the recording and the themes that would be covered. They gave their consent before their inclusion in the study. Verbal consent was witnessed and formally recorded.
Population data collection
For each participant, the following sociodemographic characteristics and psychosis history were collected: gender, age, education, minority ethnic group, being a first- or second-generation migrant, comorbidities and associated symptoms, diagnosis, duration of involvement in FEP programme, family history, medical and psychiatric history, and history of substance misuse. Participants were also asked about the presence of a specific context of onset of their episode or precipitating factors. Participants were interviewed for 60 to 180 min by A.-A.T., a female psychiatric resident trained in CAARMS interviewing and qualitative analysis. Before the beginning of the interview, a communication contract was presented, and the participant was asked for their consent for the study. During the interview, the interviewer first verified and completed sociodemographic information. Next, the participant and interviewer completed the Trauma and Life Experiences Checklist (TALE). The TALE is a 22-item scale that can be either self- or researcher-administered. It is designed for routine trauma screening in psychosis services. It includes a list of common traumatic or stressful life events, as well as an item where participants can discuss traumas not covered in the questionnaire. It has the advantage of including psychosis-specific potentially traumatic events (the experience of psychotic symptoms, hospital admission and unusual behaviours). It has moderate psychometric acceptability overall, with excellent convergent validity and reliability for sexual abuse.Reference Carr, Hardy and Fornells-Ambrojo15
The researcher-administered TALE questionnaire was not audio recorded, as this was often the first meeting with the participant and the questions were very personal. The results of the TALE were collected in a Microsoft Excel document and detailed notes were made with the participants’ answers to each question. Third, the interviewer conducted an individual qualitative interview that was audio recorded with the participant’s consent. The qualitative semi-structured interview consisted of six main questions with follow-up (see Appendix 1: semi-structured interview guide). The interview was stopped when all questions were answered, including the follow-up questions. Interview duration then varied depending on the verbatim lengths for each patient or the global speech speed. This allowed each participant to specify the context of onset, describe the positive psychotic symptoms experience, detail the characteristics of the positive psychotic symptoms and develop their opinion about the origin of the symptoms and an eventual link between trauma and life events (see Table 1). The questions exploring positive psychotic symptoms and the experience of positive psychotic symptoms were derived from the CAARMS. The CAARMS is a semi structured interview with a French validation that assesses and details positive psychotic symptoms, in particular attenuated positive psychotic symptoms.Reference Krebs, Magaud, Willard, Elkhazen, Chauchot and Gut14
Table 1 Theme 4: Thematic associations between specific positive psychotic symptoms and specific life and trauma events, with quotes from the perspective of participants
Analysis
Baseline demographic characteristics were summarised using means (for continuous data) and frequencies (for categorical data).
The qualitative part of the interviews was audio recorded, pseudonymised using a code preventing direct identification of the participants and literally transcribed. A thematic content analysis, following the approach proposed by Bardin,Reference Bardin16 had been carried out using NVivo (version 14.23.2 for Windows; Lumivero, Denver, CO, USA; https://lumivero.com/products/nvivo/ ) to get as close as possible to the participants’ discourse. A vertical and transversal analysis was carried out to categorise the transcriptions into themes and subthemes. The analysis grid followed deductive and inductive approaches. Positive psychosis symptoms coding derived from a deductive analysis based on the interview structure of the CAARMS to identify known types of positive psychotic symptoms. Next, we proceeded to an inductive analysis to identify the thematic content of each code (i.e. the content of the auditory verbal hallucinations). An initial codebook was developed based on concepts from the interview guide (i.e. from existing literature, and following the structure of the CAARMS and the TALE), and those that emerged directly from the data (i.e. from two researchers’ readings (A.-A.T. and M.M.) of 17 transcripts). The two researchers, supervised by F.H. and J.H., discussed coding choices until agreement, or a new code was developed. The codebook was subsequently finalised and all transcripts were coded. Encoding was double-blind for the first interviews and compared between A.-A.T. and M.M. to limit the subjectivity of encoding, only after A.-A.T. completed the coding of all the transcripts.
From the results of the interview and the TALE administration, the prevalence of each trauma and life events before the first episode of psychosis in the sample was calculated, except for items 15 to 18 of TALE. These items explored the traumatic experience of symptoms, the distressing behaviours possibly related to symptoms, the experiences in mental health services and other contacts with health or justice services. The prevalence of multiple exposure to each trauma and life events was also calculated. The number of trauma and life events per participant and the number of trauma and life events with multiple exposure per participant was calculated. An average age at onset and age at end of exposure to each trauma type was calculated.
For the following analyses, the participants’ answers handwritten by the examiner during the TALE completion (preceding the audio-recorded interview) and the TALE results were integrated into NVivo and coded.
We used matrix coding queries in NVivo 14 to explore the association between codes of specific types of positive psychotic symptoms and codes of specific life and trauma events. Using codes for positive psychotic symptoms and codes for traumatic event histories, each participant’s file was sorted into static sets based on either the presence or absence of that specific antecedent and that specific symptom. These sets were then subjected to matrix coding queries to reveal patterns. This enabled us to highlight preferential associations between codes for each positive psychotic symptom and codes for each life and traumatic event.
We used cluster analysis in NVivo 14 with the Jaccard coefficient, to support the identification of associations between codes for trauma and life events and other codes.Reference Hennig17 The use of cluster analysis in qualitative research is a documented method that supports thematic analysis and interpretation by revealing patterns in how codes co-occur across participants.Reference Guest and McLellan18,Reference Henry, Dymnicki, Mohatt, Allen and Kelly19 It is complementary to thematic analysis and can help to identify emerging themes or subthemes and refine thematic analysis results. This approach also improves reproducibility of the results, and is very appropriate for exploratory studies with large textual data. Among similarity measures, the Jaccard coefficient was chosen because it is commonly used when the data is qualitative and binary, such as the presence or absence of a feature. The Jaccard similarity coefficient accounts for shared occurrences of codes without inflating similarity owing to non-occurrences. The Jaccard similarity coefficient had been used to evaluate the coding similarity, i.e. it measured the proportion of participants having these two codes relative to the total number of other participants having one of these codes. The Jaccard coefficient ranges from no similarity (0) to full similarity between codes (1). According to the rules from Evans correlations, we considered correlation between 0.40 and 0.59 to be moderate, between 0.60 and 0.79 to be strong and between 0.80 and 1.0 to be very strong.Reference Evans20
Results
Description of the sample
The sample consisted of 17 individuals currently enrolled in the PEP’S FEP programme (see demographic characteristics of the sample in Table 2). Of them, nine participants were recruited from the out-patient unit and eight participants were recruited from the specialised in-patient unit for FEP. A.-A.T. knew eight of the 17 participants before the study start, from residency in the FEP programme; she also had a work history with the clinical team engaged with all the participants. This situation increased confidence between patient and interviewer and eased the interview process. Also, the use of a systematic semi-structured interview allowed reproducibility. Of the nine participants recruited from the out-patient unit, eight of them were admitted to hospitals outside of the specialised FEP programme in general psychiatric units during their first psychotic episode, and one had not been admitted to hospital. Only two participants refused to participate in the study. These two participants did not wish to have any additional interviews beyond those scheduled as part of their usual care.
Table 2 Characteristics of the participants interviewed (N = 17)
Results of the TALE questionnaire
Sixteen out of 17 participants reported past life and trauma events (see Table 3). All but one participant reported multiple trauma and life events. One participant identified the first episode of psychosis as the only trauma event in their life. The average number of trauma and life events before FEP was six. The average number of trauma and life events with multiple exposure before FEP was four. Most participants identified trauma and life events still affecting them (with a current emotional impact), in particular sexual abuse before the age of 16 years old, witnessing domestic violence, bullying, psychosis, emotional neglect, accidental events, temporary separation from a close relative, migration and emotional abuse.
Table 3 Trauma and life events cross-referenced from the Trauma and Life Experiences Checklist, interview and medical file
Results of the thematic content analysis
After conducting thematic content analysis, five major themes were identified: theme 1, types of positive psychotic symptoms; theme 2, context of onset of positive psychotic symptoms and precipitating factors of positive psychotic symptoms; theme 3, experience of positive psychotic symptoms; theme 4, thematic association between trauma and psychosis; and theme 5, causal or mediating factors of positive psychotic symptoms.
Types of positive psychotic symptoms
All participants described positive psychotic symptoms during the qualitative interview: perceptual anomalies, delusions and non-bizarre ideas, unusual thought content and disorganisation (see Box 1 for the detail of positive psychotic symptoms and the thematic content of each code).
Box 1 Theme 1: symptoms with subthemes and codes detail and number of individuals among the 17 participants interviewed for each code
Perceptual anomalies (13/17)
-
• Auditory-verbal hallucinations: content: control, body/bones, death, imperfection/failure, exclusion, injunction to arm oneself/threatens, insults/derogatory remarks, spiritual (identity of voices, relation with the person, powerful voices)
-
• Somaesthetic anomalies: heat, pain/twisting/decomposition/strangulation, electricity, sore throat/dryness, headaches, muscular pain
-
• Tactile hallucinations: drop of sweat/wet, liquefaction, gripped/touched/embrace, burn, pressure on orifices, insect
-
• Visual hallucinations: creatures/monsters, people/shadows, animals, beads of light/spiritual/symbols/cathedral vault, fire/blood/violence/red, particles
-
• Auditory hallucinations: clock noise, tik tak noise, neighbour noise, screams, hubbub
-
• Auditive distortions: auditory hyper-reactivity, greater sensitivity
-
• Visual distortions: visual hyper-responsiveness, improved visual acuity, object movement, deformation
-
• Gustative distortions: taste of paper, blood, limestone, dog food
-
• Olfactory hallucinations: failure to recognise the scent of one’s own laundry detergent or perfume, a more sensitive sense of smell
-
• Tactile distortions: embracing sensation, tactile hypersensitivity
Delusions and non-bizarre ideas (16/17)
-
• Persecutory and suspicious delusions: tracked, monitored, spied on, negatively judged, envied, life-threatened, ideas of distrust
-
• Grandiose delusions and ideas: megalomaniac ideas, ideas of controlling other people, ideas of liberation, societal ideas
-
• Mystic or esoteric delusions: bewitchment, ideas of incarnation, esoteric ideas, religious ideas, ideas of creating and end of the world
-
• Somatic delusions: pheromone release ideas, ideas of liquefaction, ideas of physical intrusion, ideas of being mute or deaf, ideas of having hot flashes, ideas about being stained, ideas of having a physical problem
-
• Nihilistic ideas: feeling of dying, sensation of near death, feeling of being dead
-
• Delusion of control: ideas of being tied down, tortured and controlled by a malefic being, ideas of being controlled by a higher being, ideas of being controlled by voices
-
• Other non-bizarre ideas: culpability ideas, ideas of infestation, nature-related ideas
Unusual thought content (12/17)
-
• Delusions of reference
-
• Perplexity: strangeness, wrong wave, astonishment, doubt, questioning
-
• Imposed thoughts and feelings: thoughts in the wrong direction, feeling imposed
-
• Intrapsychic hallucinations: self-deprecating
-
• Delusions of thought broadcasting: remarks about others, judgement
-
• Delusions of thought reading: disease, pregnancy
Disorganisation (12/17)
-
• Disorganised behaviour: abnormal clothing, screams, pathological buying, research on a deceased person, mess/damage/tearing documents, laying down in the street, boiling and cold showers/cold water, dark clothing, pathological fandom
-
• Disorganised speech: incoherent, illogical speech, unclear association
-
• Soliloquy: screams, talk, mimetic
-
• Inappropriate affect: unprovoked laughter, grimace
Negative symptoms (5/17)
Post-traumatic symptoms (7/17)
-
• Nightmares, flashback, revival, avoidance, hypervigilance
-
• Post-traumatic symptoms of positive psychotic symptoms
Other psychopathology (17/17)
-
• Depressive symptoms: depressed mood, anhedonia, ideas of incurability
-
• Sleep disturbance: insomnia, nychthemeral rhythm inversion
-
• Reduced or difficult communication
-
• Fatigue
-
• Suicidal ideas
-
• Tachypsychia
-
• Organisational difficulties
-
• Increased energy
-
• Dissociation, depersonalisation
-
• Incuria
-
• Disturbance of appetite
-
• Obsession and compulsion
The context of onset and precipitating factors of positive psychotic symptoms
All but one participant identified a specific context of onset of the FEP.
Most participants described positive psychotic symptoms as emerging in the context of trauma and life events.
Some participants described positive psychotic symptoms following post-traumatic symptoms.
The other contexts identified were travel overstimulation or climate adaptation, toxic consumption or weaning, socioeconomic or professional concern and overwork.
Experience of positive psychotic symptoms from the perspective of participants with FEP, including emotional and distressing experiences
The participants’ experience could be gathered into four subthemes: the emotions felt, the behaviours adopted by participants when experiencing positive psychotic symptoms, and the cognitive and metacognitive perturbations described by participants. Most participants described positive psychotic symptoms as a distressing experience, with fear and anxiety (see emotional responses in Box 2). They described cognitive bias, memory perturbations and metacognitive perturbations, as well as dysfunctional behavioural responses that probably maintain the distressing experience of positive psychotic symptoms (see codes, prevalence and quotations in Box 2, Theme 3: experience of positive psychotic symptoms, codes with quotes).
Box 2 Theme 3: experience of positive psychotic symptoms, codes with quotes
Emotional response (17/17)
-
• Fear or anxiety:
-
○ ‘I was afraid of being hurt’ (participant 2)
-
○ ‘I was afraid to die’ (participant 4)
-
○ ‘I put myself in a state of total panic’ (participant 7)
-
○ ‘Bad voices that put me down’, ‘I was afraid of not knowing how to defend myself… of not being safe’ (participant 14)
-
○ ‘I’d get really nervous’ (participant 15)
-
-
• Anger:
-
○ ‘Anger in fact, a lot of anger and no more control, no more control over myself in fact, at that moment I was elsewhere, I was no longer myself in fact, I was no longer how to say uh… I was another person in fact’ (participant 16)
-
-
• Sadness:
-
○ ‘I didn’t feel well, I mean I wasn’t happy, and in fact I felt mostly, I don’t know if that’s the right word, overwhelmed’ (participant 15)
-
-
• Association of fear, anger and sadness:
-
○ ‘Fear, terror, stress, anxiety, deep sadness and a feeling of helplessness that you’ve tried everything you can think of’ (participant 7)
-
○ ‘It’s stressful. So inevitably because you don’t understand, I’d say it was a surplus of stress in terms of emotion. It can also lead to a certain amount of sadness, because you don’t understand, and it’s always unpleasant to see sudden changes that you can’t control, so I’d say there was a certain amount of sadness, and I sometimes cried myself to sleep’ (participant 11)
-
○ ‘Well, it affected me… it made me sad, after all, I very rarely cry, I’m very sensitive, but it’s very rarely that I cry, but that day, it wasn’t fear, it was… it was the injustice that I… I felt. In fact, uh, I felt… well… I just felt misunderstood’ (participant 13)
-
-
• Excitement, joy:
-
○ ‘He’d tell me to look for the loophole, and it’s something that’s marked me, almost exalted me. There were all the famous singers, that’s in my hallucinations I think, all the famous singers who sang for me, but against me, who challenged me to a duel, a battle, and I had to sing, dance, and I was actually having fun. The same thing happened when, in my head, I was contacting the secret service (laughs) and I was thinking, okay you’re being very careful, I think that in real life my parents, looking at me from a distance, the doctors, must have thought, she’s talking really loudly, I don’t know (laughs) but yeah, it was upsetting, but it made me feel good, it’s weird, it really made me feel so good’ (participant 2)
-
-
• Blunted affect, difficulty to specify emotion:
-
○ ‘It’s as if I had no emotions at all since, I kept everything inside me I’d told my parents about this malaise but as soon as I saw that … they didn’t show up to uh … for my health to improve uh … I kept everything inside me’ (participant 17)
-
-
• Shame:
-
○ ‘It was like I was dying, I was ashamed of myself, and I was at my lowest ebb, that’s all…. Shame, embarrassment’ (participant 12)
-
Associated perturbations of cognitions (13/17)
-
• Memory perturbation: amnesia, association, disconnection, selective memory, over general memory
-
• Cognitive bias: attribution bias, disruption of attentional control, salience bias, interpretation bias, emotional reasoning, magical thinking, arbitrary inference
Associated metacognitions (17/17)
-
• Cognition about one’s functioning: ‘I feel like my mind has split into several pieces’ (participant 15)
-
• Cognitions about the liberating virtue of the symptoms: ‘By living through this event, it allowed me to free myself from my trauma, and maybe my brain would react better next time. (…) so when we suffer a crime, we emit pheromones to say “I’ve been raped”, for example, and we’re marked by these pheromones to indicate to the rest of the species that we’ve been raped, so to free ourselves, I’ve found nothing other than to repeat the crime in reverse on ourselves, to be the rapist and the raped, so that cancels each other out, and as a result I noticed that there were odour emanations, but very unpleasant ones for example, that’s pretty horrible to say, but I self-polished to try and free myself from it. And once I’d done that, I don’t even have words to describe how I felt after I’d done that, because I felt so good. But on the other hand, it was a horrible act and I made a slightly protective denial of it, to tell myself no, I wasn’t raping myself, but rather no, I was taking pleasure with myself, rather than telling myself that I was raping myself to free myself, I told myself more, I made love to myself, telling myself that it made me feel much better afterwards, so it was an act of love, rather than a criminal act’ (participant 1)
-
• Cognitions about self-representations
-
○ Cognition about identity and one’s existence: ‘In fact, I didn’t understand myself, I knew who I was but deep down I was nothing, in the end I was just a little speck of dust on the planet like everyone else, and the fact that I said to myself, I’m almost nothing’ (participant 2)
-
○ Negative belief about the self: ‘I didn’t particularly like myself’ (participant 2)
-
○ Body image: ‘When you’re walking down the street and you imagine you’ve got a stain etc., and I’m not talking about a stain like that, you always imagine the worst, so you can’t walk with your head held high’ (participant 12)
-
○ Increased sense of self-efficacity and accomplishment: ‘I’ve even regained my self-confidence, I’ve always had little self-confidence because of all this, and I said to myself, you can still do things on your own!’ (participant 1)
-
○ Decreased sense of self-efficacity and accomplishment, sense of failure: ‘The things I was supposed to deal with, I just couldn’t deal with them anymore’ (participant 3)
-
○ Lack of understanding/feeling lost: ‘I would say distraught, lost’ (participant 7); ‘I didn’t understand’ (participant 10); ‘I understood nothing, I was lost’ (participant 14)
-
○ Discrepancy between the self-perceived by others, the desired self or the actual self: ‘I compare myself to people and if I compare myself and I don’t like it and it doesn’t suit me, well, I’ll brood about it’ (participant 9)
-
Behavioural response (16/17)
-
• Behavioural avoidance:
-
○ Withdrawal to home: ‘I just wanted to isolate myself and that’s why I didn’t go out for the 6 months in 2022 from September to December’ (participant 12); ‘I only went out to eat and to the toilet and shower. I stayed all day with a fan and in the dark’ (participant 17); ‘I didn’t leave my house anymore’ (participant 9)
-
○ Social withdrawal: ‘I had begun to cut ties with everyone’ (participant 14); ‘In those moments I no longer saw the other people’ (participant 15)
-
-
• Repression/denial: ‘There is nothing that bothers me’ (participant 17); ‘I personally don’t mind. I’m a person, that’s all’ (participant 8)
-
• Thought suppression: ‘My family was devastated by it, even my cousins and uncles, it went everywhere, everyone knew it. And I wasn’t proud of it. I think I’ve gone through a phase where I don’t want to think about it anymore and I’ve forgotten a lot of stuff’ (participant 14)
-
• Distress endurance: ‘I was just thinking, I’ll take it as it comes and see where it takes us’ (participant 1)
-
• Ruminating and questioning: ‘It works on me; it works on my mind’ (participant 3); ‘I began to wonder about things, my analyses began’ (participant 6); ‘It was the beginning of a long reflection’ (participant 11); ‘I was thinking of ten thousand things, ten thousand solutions, um, treatments, what I could do, could I see energy therapists, were there plants that could help me with that to calm my anxieties’ (participant 7); ‘I could spend hours thinking about questions that are on my mind’ (participant 12)
-
• Distraction/suppression:
-
○ Toxic consumption: ‘I used to drink alcohol to fall asleep’ (participant 3); ‘I used drugs at home… Saturdays and Sundays, as soon as I had days off, I’d get high on … coke or alcohol’ (participant 9)
-
○ Overinvestment in studies or work: ‘I went on with my life, working every day of the week’, ‘I was congratulated at work’ (participant 9); ‘I was a bit overinvested in my studies’ (participant 15)
-
○ Compulsion: ‘checking, OCD [obsessive–compulsive disorder]’ (participant 12); ‘I had a kind of hypnotic crisis, I don’t know if you can say it, where I couldn’t get out, I was like, well, it’s also buttons that I’ve always itched since I was little, when I do it, I can’t stop, it’s horrible, and for me I call it a hypnotic moment because I’m hypnotised at the time and I can’t get rid of it, and it relieves me because it allows me to exteriorise’ (participant 2)
-
-
• Help-seeking behaviours: ‘I called my friend’ (participant 2); ‘I made sure to call ahead’ (participant 1)
-
• Activation and communication with others: ‘I wrote my book and all that’ (participant 1); ‘I wrote a sort of letter to the teacher, which is a bit strange, but I’d have to find it in my file, I think it’s there, where I explained a bit what I thought about the world’ (participant 5); ‘I wrote a letter’ (participant 6)
Factors influencing symptoms
-
• Toxic consumption
-
• Sleep
-
• Anxiety, stress
-
• Behaviours adopted
-
• Loneliness
-
• No factors identified
Thematic associations between specific positive psychotic symptoms and specific life and trauma events
Numerous direct and indirect thematic associations were identified, but only those made by the participants were reported (see Table 1, Theme 4: thematic associations between specific positive psychotic symptoms and specific life and trauma events with quotes from the perspective of participants).
Causal or mediating factors of positive psychotic symptoms
The majority of participants developed causal theories about the first episode of psychosis and positive psychotic symptoms. Only three participants did not develop causal theories about the first episode of psychosis, and all of them had negative symptoms in the foreground at the time of the interview.
Most participants had theories including life and trauma events as causal factors of positive psychotic symptoms (see Box 3. Theme 5: participants’ causal theories of positive psychotic symptoms, with number and quotes). Of them, one participant described the positive psychotic symptoms to avoid exposure to ongoing trauma events. Three participants mentioned their life and trauma events, but did not know if they were causal factors. One participant who migrated to France to study and work, identified the recent separation from his family as traumatic, but made no connection with the first episode of psychosis. Four participants described psychological processes as mediating factors: negative belief about others because of trauma events, repetitive questioning or attribution bias.
Box 3 Theme 5: Participants’ causal theories of positive psychotic symptoms with number and quotes
Participants were asked their opinions on a narrative interview discussing the link between their positive psychotic symptoms and trauma and life events.
The majority of participants identified the interview as relevant and helpful. The arguments put forward were to understand psychosis considering their life history, because the two of them fit together, to integrate psychosis and construct a narrative, to accept the disorder and to turn the experience into something positive. One of them specified that it is more helpful shortly after the first episode, because after they did not appreciate talking more about it. One of them found it helpful but could not tell why.
One participant identified a link with the context of onset of the first episode of psychosis and explained that they did not need help to develop it.
Four participants did not think there was a link between trauma and psychosis, and thus did not see the relevance of such an interview.
Exploration of the association between specific positive symptoms and specific trauma and life events drawn from the matrix coding queries
For this phase, the participants’ answers handwritten by the examiner during the TALE and the TALE results were integrated into NVivo and coded.
Exploratory associations between specific positive psychotic symptoms and specific trauma and life events have been drawn from the matrix coding queries (see Table 4). The main associations were:
-
(a) Tactile perceptual anomalies, bewitchment or incarnation ideas, thoughts/images/acts imposed on self, auditory-verbal hallucinations, combination of megalomaniac and persecutory ideations and sexual abuse before 16 years old. Three participants had all of these symptoms combined; all of these participants reported sexual abuse. The association between all of these symptoms was only found in patients reporting sexual abuse;
-
(b) Combination of ideas of grandiosity and control and ideas of being prey to others, disorganised behaviour and permanent loss of caregiver and physical neglect;
-
(c) Soliloquy and witnessing domestic violence;
-
(d) Persecutory ideas and bullying, intimidation and emotional neglect.
Exploratory associations between trauma and life events codes, positive psychotic symptoms codes, and other codes (emotion, cognitive and metacognitive, context of onset, thematic association, causal theories) have been made with cluster analysis (see Table 5). Only trauma and life codes present in at least three participants are shown. We found that intimidation or bullying was correlated with megalomaniac delusion; emotional abuse with disorganised behaviour; emotional neglect with ideas of reference, grandiose delusions, suspiciousness and paranoid delusions; childhood sexual abuse with tactile hallucinations, mystic or esoteric delusions, ideas of incarnation, auditory verbal hallucinations, megalomaniac delusions and delusion of control; and witnessing domestic violence with soliloquy.
Table 4 Other exploratory associations between specific positive psychotic symptoms and specific trauma and life events drawn from the matrix coding queries of the qualitative analysis
Table 5 Cluster analysis to explore similarity and dissimilarity between codes of trauma and life events, and other codes (with Jaccard coefficient superior to 0.5 indicated)
TALE, Trauma and Life Experiences Checklist.
Discussion
This study investigated the experience of positive psychotic symptoms and the association between positive psychotic symptoms, life and trauma events, from the perspective of people who have experienced a first episode of psychosis.
To our knowledge, this is the first study to explore the experience of FEP among patients in France. Our qualitative findings give important insights into the experience of FEP and individual beliefs about the causes of their distress. We asked patients for their opinion on a possible association with their life and trauma events and the relevance of such interviews in routine care. Our study also gives exploratory data on the period of exposure to life and trauma events in this population, and on the association between specific positive psychotic symptoms and specific trauma and life events. Most of the participants were exposed to cumulative trauma.
Our study revealed that the experience of positive psychotic symptoms was associated with emotional distress for most participants; some subtypes of positive psychotic symptoms were preferentially associated with trauma events; there were thematic association between trauma events and the symptoms, and, for some participants, positive psychotic symptoms were associated with (or preceded) dissociation and post-traumatic symptoms.
Association between specific types of positive psychotic symptoms and trauma and life events, and characteristics of positive psychotic symptoms
The association found between tactile anomalies, mystic or esoteric ideations, thoughts/images/acts imposed on self, auditory verbal hallucinations, combination of megalomaniac and persecutory ideations, and sexual abuse before 16 years of age is consistent with some other studies.Reference Read, Agar, Argyle and Aderhold21,Reference Daalman, Diederen, Derks, Van Lutterveld, Kahn and Sommer22
Ideas of controlling others and being prey to others and behavioural disorganisation were associated with permanent loss of caregiver, physical neglect and discrimination. In a transdiagnostic sample, paranoia and grandiosity were positively related to interpersonal trauma and unrelated to non-interpersonal trauma; these relationships were significantly mediated by impaired sleep.Reference Herms, Bolbecker and Wisner23 The association between disorganised behaviour and trauma was not clear in the literature.
The association between persecutory delusions, bullying and intimidation, discrimination and emotional neglect is in agreement with other studies.Reference Catone, Marwaha, Kuipers, Lennox, Freeman and Bebbington24,Reference Wickham and Bentall25
In our study, soliloquy was associated with witnessing violence at home. In some participants, soliloquy was identified in their medical file, and other patients reported it themselves. For the one who reported it, it is difficult to establish if it was soliloquy or hallucination of soliloquy.Reference Kobayashi and Kato26 The association between soliloquy and domestic violence was not found in the literature.
Thematic associations were found between trauma and life events and positive psychotics symptoms such as hallucinations (olfactory, tactile, auditory verbal, somaesthetic) and delusions (death threat, suspiciousness, persecutory delusions, delusion of control, spiritual ideas). Thematic associations were also found between post-traumatic symptoms and positive psychotic symptoms. This is in agreement with other studies.Reference Gerges, Haddad, Daoud, Tarabay, Kossaify and Haddad27,Reference Offen, Thomas and Waller28
In our study, delusion of life threat with altered behaviour to protect oneself was presented by two participants exposed to deliberate life threat. These participants reported both physical abuse, emotional abuse and emotional neglect. In a cross-sectional study, high perceived life threat was associated with post-traumatic stress disorder.Reference Heir, Blix and Knatten29 We hypothesise that exposure to life threats induced deliberately might favour delusion of life threat.
Time of exposure and cumulative exposure
In the study, some life and trauma events were preferentially associated with subtypes of positive psychotic symptoms.
The period of exposure to domestic violence in the population was between 5 and 13 years old, which corresponds to the suggested period of brain sensitivity to childhood trauma and adversity.Reference Teicher, Samson, Anderson and Ohashi30 In our sample, exposure to domestic violence was mainly exposure to verbal violence. We hypothesise that exposure to domestic verbal violence could have an impact on regions involved in the control of own-voice soliloquy, the inhibition of motor commands and first-person and third-person perspective.Reference Grandchamp, Rapin, Perrone-Bertolotti, Pichat, Haldin and Cousin31
The period of exposure to childhood sexual abuse was between 10 and 12 years old. The pathway of genesis to a psychiatric disorder in response to childhood sexual abuse is complex and determine by multiple factors, such as genetic, epigenetic, neurobiological changes, neurochemical, synaptic changes and effects of neuroendocrine axes in response to stress.Reference Shrivastava, Karia, Sonavane and De Sousa32 In our sample, exposure to sexual abuse before the age of 16 years was associated with memory perturbation, imposed visual mental imagery, anger, tactile and auditory-verbal hallucinations, persecutory and grandiose ideas, and delusion of control. Our finding is compatible with the hippocampal abnormalities found in neuroimaging studies,Reference Pederson, Maurer, Kaminski, Zander, Peters and Stokes-Crowe33 possibly switching off control over the limbic system, which then exacerbates emotional and post-traumatic stress disorder-related symptomsReference McCrory, De Brito and Viding34 in response to sexual abuse. The symptoms of our participants who reported sexual abuse are congruent with the effect on the cerebellar vermis, a structure intricately linked to multisensory integration and limbic activation.Reference Schiffer, Peschel, Paul, Gizewski, Forsting and Leygraf35
Participants’ narrative and causal theories
Most participants developed causal theories of positive psychotic symptoms including life and trauma events. Their theories including trauma and life events were compatible with theoretical models proposed in the literature concerning the association between trauma and psychosis: stress sensitivity,Reference Lardinois, Lataster, Mengelers, Van Os and Myin-Germeys36 dissociation,Reference Braehler, Valiquette, Holowka, Malla, Joober and Ciampi37 self-disgust as a mediating factor,Reference Simpson, Helliwell, Varese and Powell38 social pathways such as attachment disruption and social rank,Reference Heriot-Maitland, Wykes and Peters39 and negative schemas about others and self.Reference Humphrey, Berry, Degnan and Bucci40
Most participants found a narrative interview about the link between trauma and life events and positive psychotic symptoms relevant and helpful. Such interviews allow individuals to be validated in their emotional experience, and to adopt an agent posture in which they develop an understanding of their experience in their personal lives and integrate their life experiences.
Strengths and limitations
The study gains its strengths from the combination of in-depth qualitative analysis and exploratory cluster and matrix coding analyses. To our knowledge, this is the first study to apply this method to an FEP population.
All participants lived in Greater Lyon, which is mainly urban, and were involved in a FEP programme either in the in-hospital unit or in the out-patient programme with multidisciplinary care. Data may not reflect the views of people in other areas of France and outside FEP services. We recognise that several analyses are based on a small number of observations, which limits our ability to draw general conclusions. In particular, the data linking specific positive symptoms to certain types of trauma are based only on three participants. However, we felt it relevant to mention them, as they may inspire more systematic and quantitative explorations for this type of association.
Future research should consider testing the hypothesis derived from the exploratory cluster analysis and matrix coding analysis, and explore the neuroanatomy and the cognitive mechanisms underlying or mediating trauma and life events and positive psychotic symptoms.
In conclusion, our participants described a distressing experience of positive psychotic symptoms and multiple causal theories of their symptoms, often linking it with life and trauma events. Our findings point to the potential benefits of interviews exploring their symptoms considering their life and trauma events.
Data availability
The data that supports the findings of this study are available from the corresponding author, F.H., upon reasonable request.
Acknowledgement
The authors thank the PEP’S medical and case manager staff for their help in the recruitment process.
Author contributions
A.-A.T. contributed to data curation and analysis, and was responsible for writing the original draft and subsequent revisions. F.H. was involved in conceptualisation, supervision and contributed to writing through manuscript review and editing. J.H. contributed to methodology development, supervision and manuscript review. M.M. was involved in data analysis and methodology.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.
Declaration of interest
None.
Appendix 1. Semi-structured interview guide
Open access
eLetters
No eLetters have been published for this article.