WHAT IS ALREADY KNOWN ON THIS TOPIC
-
Basic self-disturbance (BSD) refers to various forms of anomalous self-experiences (ASEs) characterised by self-alienation and loss of feeling of natural presence in the world.
-
BSD is a core marker for schizophrenia and is also suggested as a prepsychotic vulnerability marker.
-
Studies on adult populations have demonstrated that BSD is temporally stable (trait-phenomenon).
-
As for clinical use, too little is known about the temporal development and stability of BSD in early prepsychotic phases in childhood and adolescence.
WHAT THIS STUDY ADDS
-
Time-series data collected prospectively during 6 months from the smartphones of 27 adolescents revealed relatively stable levels of ASEs within persons and across the whole group, irrespective of time periods and baseline.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE AND POLICY
-
The demonstrated relative stability of BSD in adolescents indicates that BSD may be considered a trait-phenomenon even across younger age groups.
-
Considering the continuous and everyday experience of BSD and its predictive value for psychosis development, ASEs should be targeted and monitored to the same extent as other prepsychotic features.
Background
Basic self-disturbance (BSD), also called anomalous self-experience (ASE) or self-disorder, is a core phenotypic marker for schizophrenia spectrum disorders (SSD) and evidence suggests that it is a prepsychotic vulnerability marker.1 Intriguing experiences of self-alienation, feelings of not fully ‘inhabiting’ oneself and not being the owner of one’s thoughts, experiences, actions or body sensations are fundamental characteristics. BSD is the overarching concept for these distinctive experiences which are also referred to as ASEs. The ASE concept was introduced and defined through the publication of the phenomenological checklist Examination of Anomalous Self-Experience (EASE)2 in order to specifically explore a range of the various forms of BSD referred to as ASEs. In this study, we will use the two concepts accordingly.
The occurrence and distribution of BSD in SSD, other psychotic and psychiatric diagnoses and healthy controls have been systematically mapped in several studies of adults, all showing a consistent and significant aggregation of BSD in SSD. Also, in samples with SSD studies have found BSD to be consistently and strongly associated with social dysfunction,3 depressive symptoms and suicidality.4 5 A strong association between negative symptoms and BSD has been found in individuals considered to be at a clinical high-risk for psychosis (CHR-P).6 The predictive diagnostic value and the temporal stability of BSD have been examined in studies spanning different periods of time (1–7.5 years).7–9 These studies have also found high baseline scores of BSD to predict a rediagnosis with SSD in patients who were initially diagnosed outside the spectrum. Also, high baseline scores of BSD predicted transition to a psychotic disorder in patients who initially were classified as CHR-P (1.5-year follow-up). Hence, BSD is at present largely considered as trait-like. Studies have documented BSD to emerge during both childhood and adolescence but there is limited evidence on the developmental course and characteristics of BSD in prepsychotic phases. To our knowledge, only two studies have investigated BSD prospectively in individuals with a mean age under 18. In a 7-year follow-up study, Koren et al8 found SSD diagnoses to be significantly predicted by BSD in adolescence. The second, a cross-sectional study10 explored the distribution of BSD in adolescent help-seekers (age range 14–18) and tested the specificity of BSD with respect to the severity of diagnostic staging between early onset schizophrenia-spectrum psychosis (EOP), CHR and clinical help seeking controls (CHSC). The EASE total score increased the risk of belonging to the more severe diagnostic stages, that is, CHR and EOP versus CHSC (as reference class) and allowed the correct reclassification of the 75% of the sample.
BSD course and stability: further research in adolescents
Most studies investigating the stability of ASEs have used the EASE interview on adults performed at two time points with long time intervals from baseline to follow-up. Adolescence is marked by physical, emotional and social changes where rapid shifts in mood, unstable self-esteem and challenges in social relations are typical. Early psychopathology symptoms are often transient and dynamic before crystallising into a syndrome fitting some operational diagnostic categories such as the Diagnostic and Statistical Manual of Mental Disorders (DSM) or International Classification of Diseases (ICD). To be clinically useful, we need to know how ASEs behave at these transitional phases of adolescence in relation to their trait or state characteristics (or both), their temporal development over shorter and longer time spans and their potentially different temporal trajectories. To our knowledge, there is no study conducted to date on these early characteristics.
Objective
The aim of this study is to provide new insight into the course of ASEs in adolescents identified to be at high risk for developing psychosis. We pose these research questions (RQ):
RQ 1: How stable are prototypical EASE items at the level of subjects and at the level of single ASEs:
-
Over the course of six months (long-term)?
-
Over the course of single days and weeks?
RQ 2: Is the EASE total score at baseline associated with the stability and level of intensity of ASEs over the (prospective) course of 6 months?
Methods
Design
This is a prospective naturalistic multi-case study examining the different types and individual courses of ASEs in daily life at different time intervals during 6 months, in 12–19 year-old help-seeking adolescents with CHR status. We used a mixed-method design combining (1) qualitative data from full EASE interviews and (2) quantitative data from repeated samplings on a smartphone application with registrations of the frequency and intensity of personally selected ASE phenomena in an everyday context.
Participant recruitment
Patients were recruited from specialised outpatient units from Children and Adolescent Mental Health Services in South-Eastern Norway. Clinicians were encouraged in informational meetings by brochures and by mail correspondence to ask patients 12–18 years at clinically suspected risk of psychosis to participate in the study. Exclusion criteria were established psychosis, intellectual impairment, neurological disorders, current antipsychotic treatment (current or for≥4 weeks lifetime, equivalent to a dose of≥5 mg olanzapine per day), clearly substance-induced CHR-symptoms or not being fluent in the Norwegian language.
Participation in the study was founded on the condition of informed consent (for patients below 16 years both parents consented as well) before taking part in the study.
Inclusion procedure
As a screening for referral, we applied an extended version of the Prodromal Questionnaire 16-item version (PQ-16)11 supplemented with four ASE questions adapted from the EASE manual by one of its main authors PM. These ASE items were composite (ie, combining two or more closely related EASE items into one common question), selected on the basis of particularly relevant, prototypical and frequently reported EASE items in six studies12–17 as well as practical clinical and theoretical considerations. Clinically referred participants who endorsed six or more items on the PQ-16 part of the prodromal screening and additionally asserted at least one clearly positive answer to the four ASE questions continued to the Structured Interview for Prodromal Syndromes (SIPS).18 If the participant confirmed a CHR status, she/he continued to a full EASE interview. Transcripts from the full EASE interviews were read and reconfirmed by each participant whereupon she/he chose three personally significant and well-manifested ASEs formulated as statements to register on the phone in accordance with the experience sampling method (ESM)-schedule. The ESM questionnaire was installed by an encrypted and secure solution on a smartphone application specifically developed for our study. By referring to each of the three verbatim ASE statement, the participant was asked at each prompt; ‘How intense is this experience for you at this exact moment?’ Levels of intensity and distress for each of the three personal ASEs were marked on a vertical scroll bar on a seven-point Likert scale (0=not at all, 6=very much).
Diagnoses
To get a real-world, broad and clinically relevant picture of our patient group, we chose to report the principal clinical diagnoses (more than one, if applicable) achieved through all-data consensus assessments based on the ICD-10 in multidisciplinary clinical teams at the sites.
Measures
The Structured Interview for Prodromal Syndromes
Patients fulfilling the Criteria of Psychosis-risk Syndromes in the SIPS were included. Inter-rater reliability (IRR) was tested by comparing scores on nine case vignettes with the final scores of raters from the North American Prodrome Longitudinal Study. CHR- status agreement was 100% and the Scale of Psychosis-risk Symptoms (SOPS) scores IRR was excellent with an intra class correlation coefficent: 0.95, 95 % CI (0.96, 0.99), two-way mixed effects model, absolute agreement).
Examination of Anomalous Self-Experience
The EASE interview consists of 57 items divided into five rational domains: (1) cognition and stream of consciousness, (2) self-awareness and presence, (3) bodily experiences, (4) demarcation/transitivism and (5) existential reorientation. We rated the EASE based on the previous 6 months at baseline and at follow-up so that the period corresponded to the period for the ESM registrations. EASE has very good to excellent internal consistency and IRR.19 The interviews were performed conversationally, that is, the patients were asked to describe in detail in his/her own words, his/her experiences for each item and to give concrete examples. All EASE interviews were conducted (baseline and follow-up) by LB, videotaped and after scoring they were all consecutively evaluated, discussed and confirmed/disconfirmed in close dialogue with PM (one of the main authors of the EASE) as part of continuous quality check and supervision.
Sample and the ESM-schedule
Since our study was novel and the first of its kind and the main focus of this study was explorative rather than confirmatory, we were not able to estimate an adequate sample size based on prior research. Whereas qualitative research seeks detailed, in-depth information from individuals usually small samples of people (N=5–15) are required whereas sample size in ESM research requires larger samples in order to test hypotheses in representative samples. Sample size in ESM is defined as (1) the number of participants (macro level) and (2) the number of data points (micro level). We followed the 30/30 rule recommended by Kreft and Leeuw20—to include a minimum N of 30 at the macro level (our participant level) and a minimum n of 30 at the micro level (our within-subject repeated measure level). This leads to a desired total sample size of 900 (product sum of N × n). At the macro level, we included 30 CHR-adolescents (60% females, aged 15–18). First, in the 1 week ESM schedule, the application was set up to emit a signal at randomly selected time points between 7:30 and 22:30 (15 hours) within mean intervals of 90 min (range: 15 min to 3 hours) 10 times a day on seven consecutive days. Second, in the sixth-month schedule, the application was set up to emit a signal at randomly selected time points within the same 1 day interval frame but this time only during one single day (randomly allocated) every other week for 6 months/24 weeks. The total multilevel sample sizes (the total amount of measurements) for the two schedules were estimated to be 2100 and 3600, respectively.
Statistical analyses
The three personal statements of ASEs referring to single EASE-items (ASEs) from each participant had an index number 1–3. Verbatim statements expressing these three ASEs of each participant constitute the qualitative raw data in this study. Interpretative phenomenological and linguistic analyses of the same statements are reported in a previous qualitative study.21 In the present paper, for transparency and representativeness of these distinct subjective experiences, we listed all the unique statements. For data on intensity and stability scores during 6 months (RQ 1a.), we used the mean intensity scores and SD for each statement. We also calculated a mean of SD which was the mean of the SDs of the three ASEs for each participant. Further, for data on the level of individuals during time courses of (1) hour-by-hour, (2) between days, (3) between weeks and (4) months (RQ 1b.), we used, for practical and feasibility reasons, the whole group’s most frequently reported EASE-items (the same two items for all). To graphically/visually present the courses of this, we used ‘Chart builder’ with time of registration on the x-axis and mean daily intensity on the y-axis. The relationship between the EASE total score at baseline and stability over 6 months (RQ 2) was analysed by comparing mean intensity scores (across the three personal statements) between the half of participants with the lowest baseline EASE total score and the other half of the participants with the highest score (with one more participant in the low score group). The data were checked for eligibility concerning requirements for the tests used. Calculations were done by independent-sample t-tests of means. A p value of<0.05 indicated statistical significance. SPSS V.25 was used for all statistical analyses.
Findings
Of the 30 participants recruited, 27 (90%) completed a sufficient amount of smartphone replies to be included in the analyses and clinical follow-up data was possible to obtain for 23 participants. The participants had a mean age of 16.1 years and 59.3% of them were female (table 1). At follow-up, 16 persons out of the 23 (69.6%) experienced reduced SIPS symptoms, 5 patients (21.7 %) did no longer fulfil the CHR criteria and 2 persons (8.7 %) converted to diagnoses within the F20–29 categories (schizophrenia, schizotypal and delusional disorders). Notably, among the 23 patients interviewed at follow-up, there was a considerable decline in EASE by dichotomised scores (binary scores on each item where a 0-score implies that ASEs is absent and a 1-score implies that the ASE is present) from 16.4 at baseline to 11.3 at follow-up.
Sociodemographic and clinical characteristics
The qualitative (explorative) and quantitative findings are presented in different ways that complement and illuminate each other. Despite the various formulations and descriptions, they express atmospheres of unreality, experiential detachment, identity confusion and basic existential insecurity typical for ASEs, that is,: ‘My thoughts seem unreal and strange’, ‘I need to touch myself to know if I exist’, ‘I don’t feel present in the world’.
RQ 1a. How stable are ASEs during the course of 6 months (long-term)?
In the full 6-month schedule, the participants responded to 1741 prompts (total replies for all participants altogether (mean=65)) which is a response rate of 33.9%. At each prompt the participant registered the intensity of her/his three personal ASEs amounting to a total of 5223 (1741 × 3) unique reports on single EASE items/ASEs. The EASE consists of items divided into five domains. Most of the participants selected EASE-items/ASEs from domains (1) (cognition and stream of consciousness) and (2) (self-awareness and presence). 19 ASE statements out of the total of 81 statements were selected from domains 1 and 2 in the EASE. Out of the 27 participants, 18 of them (67% of the 27 participants) selected EASE-items/ASEs from these two domains. In table 2, mean intensity scores and SD for the full 6-month period for all 81 statements are listed. Considering the whole group (N=27), the single item mean intensity scores (0=not at all, 6=very much) spanned almost over the full range of 0–6 (0.07–5.91) while SD ranged from 0.27 to 2.47. Further, the mean of SDs was calculated from the three SDs for each of the three ASEs for each participant. This mean of SDs ranges from 0.54 to 2.13 (see also table 2, right column) with a mean variability of 1.25 SD for the whole group demonstrating a low level of variability in the intensity of ASEs in this sample. Of the total sample, 95% of the patients had a mean variability of less than 2.05 points on the 0–6 Likert scale (online supplemental table 2).
ASE intensity and variability among those with lower versus higher EASE total scores at baseline
RQ 1b. At an individual level, how stable are ASEs during single days and weeks (short-term)?
To be able to further demonstrate qualitatively and visualise the individuality of separate ASE courses more directly, within-subject and between-subjects, we extracted the two, by far, most frequently reported EASE items for illustrative analyses. Item 2.1; Diminished Sense of Basic Self and item 2.4; Diminished Sense of Presence were selected as two of the most significant and disturbing experiences for several participants in the group, generally also seen as prototypical core ASE features.12–17 Together, these two items were represented by 19 statements (a quarter of the 81 statements) deriving from 18 participants (67% of the 27 participants). Figure 1 shows the individual variability from this subgroup.
Detailed courses (hour-by-hour, day-by-day, week-by-week) for the two ‘core’ items in the reports from single participants. The hour-by-hour graph, for each item, shows the reported intensities (Y-axis) at the first five prompts (X-axis) during the first day of the ESM-registrations. The day-by-day graphs present the reported mean intensity (Y-axis) for each day, in seven consecutive days (X-axis) in the first week of the ESM-registrations. The week-by-week graphs present the reported mean intensity (Y-axis) during one single day, every other week, for 12 weeks (the 6-month period). EASE, Examination of Anomalous Self-Experience; ESM, experience sampling method.
A procedure for a required minimum and maximum number of replies was established for reliability purposes so that calculations were based on the same number of prompts for each participant. Detailed courses (hour-by-hour, day-by-day, week-by-week) for these two ‘core’ items respectively are represented in figure 1.
Both items, 2.1 and 2.4, demonstrate a certain degree of individual variability in ASE intensity across all three periods. To get an even clearer picture of the variations, we therefore wanted to see if variability could be related to where on the severity scale (between 0 and 7) the individual item was mainly situated.
RQ 2. Is EASE total score at baseline associated with intensity or variability of the selected ASEs over the prospective course of 6 months?
Among the half of the participants with the lowest EASE total scores at baseline, the mean level of (the personally selected) ASE intensity was significantly lower over the full follow-up period than for those with the highest baseline EASE total scores (mean 2.42 vs 3.42, p=0.046, table 2). The variability of the intensity scores represented by the mean of the three SDs (across the three statements) for each participant did however not differ significantly between the two groups calculated as group means (1.25 vs 1.24, p=0. 949).
Discussion
In this study, we aimed to take a first step towards investigating prospectively the temporal course of early BSD experiences in adolescents at enhanced risk of psychosis, during short-term and long-term, by using intensive time-series data. The results revealed overall considerable stability in the level of BSD intensity both within-subjects and between-subjects irrespective of time periods and EASE total scores at baseline.
The trait character of BSD in adolescents
This study demonstrates that ASEs for the most part in the daily life of adolescents do not like ‘on/off’ experiences with sudden shifts from one moment to the next. Rather, once they occur, they seem to be present to a certain extent most of the time demonstrating the distinct character of BSD. Apart from the temporal stability of BSD, there are other aspects of BSD also suggesting its trait character including the ‘hyper-aggregation’ (high level of selective occurrence) of ASEs in SSDs and the intrinsic heterogeneity of ASEs, that is, the inter-relation of the particular items as measured by the EASE resulting in its clinical manifestation as a particular Gestalt rather than appearing as separate and well-delineated symptoms.22 23 Our findings support the trait perspective by shedding light on two contrasting levels of selfhood: The ‘narrative’ level and the ‘pre-reflective’ level. Fluctuating identity feelings and unstable self-image are considered part of normal adolescent development manifesting itself as instability at the narrative level of selfhood. BSD phenomena, however, represent the most fundamental disturbances, that is, at the basic, tacit level of selfhood implying a far more stable prereflexive, preverbal disturbance. The relative stability of ASEs demonstrated in this study may indicate that disturbances at the prereflective level of selfhood are far more stable and encompassing than identity confusion in ordinary teenage development.24 25 These results add evidence for BSD as trait-like features across age. Further studies are warranted, on different mechanisms that may bring some adolescents from ASEs into SSD while in other individuals ASE is tapered off to insignificant levels or ‘remission’. Such studies might also compare ASEs with other potentially fluctuating factors such as emotions, self-esteem and contextual factors such as social relations and different physical settings.
Systematic phenomenology by means of the ESM
The use of smartphone notifications turned out to be a feasible and practical way of collecting data on ASEs from young persons compared with long interviews in clinical settings (the average duration of an EASE interview is approximately 90 min) with the danger of high recall bias in this age group. However, a major challenge in BSD research is to develop mixed methods (qualitative/quantitative) which would not diminish the value of phenomenological data obtained with the EASE interview but rather enable higher levels of generalisability. ESM is often called a ‘systematic phenomenology’, combining ‘pure phenomenology’ (ie, the exploration of subjective experiences) with the use of conventional empirical tools. A unique element of this study is the self-selected individual verbatim statements extracted from the EASE interviews (table 2). These data constitute exactly the kind of phenomenologically ‘pure’ data securing the highest possible degree of authenticity and personal ownership of the data. The quantitative part of the study represented by the repeated measures of the ASEs is supplementary to the phenomenological perspective in terms of temporal stability and provides stronger evidence of the trait-character of BSD. A major concern in ESM studies is that this method may alter the behaviour, thoughts and emotions of participants affecting not only the validity of data but also the clinical symptoms in the participants. Previous studies have mixed results, some studies have reported reactive changes26 while others have failed to find such effects.27 We did not explore this in our study and these effects cannot be excluded.
Clinical implications
Despite the rapidly increasing research into BSD over the last 20 years, clinical interventions systematically targeting BSD are still lacking. An outline of a pragmatic seven-step treatment module has just recently been proposed28 but is in an early phase and further evaluations of implementation and clinical trials of this intervention remains. Clinical staging individualised models for clinical practice are now widely acknowledged for CHR conditions. Considering the predictive value of BSD for psychosis development and the continuous but individually fluctuating experiences of these distressing phenomena reported by affected adolescents in this study, we propose that ASEs should be targeted to the same extent as psychotic symptoms in CHR conditions. A common experience reported by clinicians conducting EASE interviews is that patients express relief and gratitude when they talk with someone who is familiar with these strange and intriguing phenomena and can help them to verbalise and thus get a deeper understanding and awareness of their experiences. However, ASEs are not isolated phenomena detached from contextual factors. In addition, one must address the potential contextual mechanisms that maintain or reduce the presence of ASEs in individuals. Thus, ‘BSD-treatment’ should be targeted by a flexible practice. For individuals with high instability in particular, the ESM could be one of the methods to identify and monitor potential patterns, other clinical symptoms and contextual factors (social settings, activities, mood) as potential triggers for ASEs. Such information is valuable to get a detailed picture of the patient’s lived life (which is often missed when conducting interviews in clinical settings).
Strengths and limitations
This study has several limitations. First, the limited response rate in this study (33.9%), even though expected, may have had an impact on our findings. We did not analyse factors associated with the unanswered prompts such as technological problems, skipping prompts in certain settings or times of the day. Missing data is an acknowledged limitation in ESM studies often associated with factors such as long registration periods and diagnoses such as psychosis. Second, and despite the monofactorial structure of the EASE, the extraction of three (personal) core items to report from each participant cannot replace the clinical information obtained from a full EASE interview. Items of ASEs other than the three initially selected, could have been as relevant, even more, for the participants over time. This selection of items is nevertheless necessary in a study like this for practical reasons. Finally, ‘response bias’ cannot be excluded in this study, that is, the adolescents may have been repeating similar levels of BSD at many prompts (set responses) due to response fatigue or to save time.
Data availability statement
Data are available upon reasonable request. Anonymised data are available on reasonable request to the corresponding author. Data that could identify participants will not be provided.
Ethics statements
Patient consent for publication
Ethics approval
The study was approved by the Regional Committee of Medical and Health Research Ethics (Id. no. 2016/1758c) in Norway. Participants gave informed consent to participate in the study before taking part.
