Although insight in schizophrenia spectrum disorders (SSD) has been associated with positive outcomes, the effect size of previous treatments on insight has been relatively small to date. The metacognitive basis of insight suggests that metacognitive training (MCT) may improve insight and clinical outcomes in SSD, although this remains to be established.

This single-center, assessor-blind, parallel-group, randomised clinical trial (RCT) aims to investigate the efficacy of MCT for improving insight (primary outcome), including clinical and cognitive insight, which will be measured by the Schedule for Assessment of Insight (Expanded version) (SAI-E) and the Beck Cognitive Insight Scale (BCIS), respectively, in (at least) n=126 outpatients with SSD at three points in time: i) at baseline (Time 0); ii) after treatment (Time 1) and iii) at 1-year follow-up (Time 2). SSD patients receiving MCT and controls attending a non-intervention support group will be compared on insight level changes and several clinical and cognitive secondary outcomes after treatment and at follow-up, whilst adjusting for baseline data. Ecological momentary assessment (EMA) will be piloted to assess functioning in a subsample of participants.

This will be the first RCT testing the effect of group MCT on multiple insight dimensions (as primary outcome) in a sample of unselected patients with SSD, including several secondary clinically relevant outcomes, namely symptom severity, functioning, which will also be evaluated with EMA, hospitalizations and suicidal behaviour.

INTRODUCTION:

Schizophrenia and related disorders remain associated with relatively poor psychosocial outcomes. Impaired insight has been reported to be the strongest predictor of this poor outcome in psychotic disorders. However, the effect size of previous treatments on insight changes in psychotic disorders has been relatively small to date, which may have been the result of not tackling the actual underpinnings of insight in psychosis.

Several theories have been proposed to explain what underlies 'lack of insight' in schizophrenia spectrum disorders. First, lack of insight could be viewed as having a function in terms of being protective or preserving self-esteem, i.e., a denial mechanism. Second, lack of insight may also be considered as a primary symptom of the disorder. Third, the neurocognitive basis of insight was in part supported by a meta-analysis, which also suggested that other variables, namely metacognition, which was defined as 'the ability to think of one's and others' thinking', may affect insight. Indeed, patients with schizophrenia have been reported to show metacognitive deficits and poorer metacognitive performance is linked with impaired insight in schizophrenia. Metacognitive training (MCT) may therefore improve insight, which should also have a positive impact on clinical outcomes, although no previous randomized clinical trials (RCT) have investigated this.

MCT was first developed by Steffen Moritz and Todd Woodward and is available at no cost at: http://www.uke.de./mct. Since then, one systematic review and four meta-analyses have replicated the positive effects of MCT on positive symptoms, particularly delusions, when compared with 'treatment as usual' (TAU), although one meta-analysis failed to show such an association. However, to the best of our knowledge no definitive randomised clinical trial (RCT) using MCT has considered insight as primary outcome to date, although three previous RCTs found other non-MCT metacognitively oriented therapies to improve insight in first-episode psychosis and schizophrenia.

Clinical and cognitive insight are different, albeit related, concepts. There has been a growing interest in clinical insight in psychosis since the multidimensional model of insight proposed by David, which encompasses three different, albeit overlapping, dimensions - illness recognition, symptom relabelling and treatment compliance - and has been consistently replicated ever since. Multidimensional measurement scales, such as the Scale of Uawareness of Mental Disorder (SUMD) and the Schedule for Assessment of Insight (SAI-E), were also devised for research. Cognitive insight is a core metacognitive domain which refers to the person's ability to evaluate and correct his/her own distorted beliefs and misinterpretations (self-reflectiveness) and the tendency to overconfidence in one's conclusions (self-certainty).

The main aim of this RCT is to test whether MCT can improve insight, including clinical and cognitive insight, in patients with schizophrenia over a 1-year follow-up. As secondary aims, the effect of MCT-related insight changes on clinical outcomes, including symptomatic severity, hospitalizations, suicidal behaviour and psychosocial Functioning, will be investigated. In addition, the investigators will pilot the use of ecological momentary assessment (EMA) via two web-based applications - www.MEmind.net and the Evidence-Based Behaviour platform eB2 app - to measure functioning in a subsample of participants.

Specifically, the following five hypotheses are to be tested: i) that MCT will result in higher cognitive and clinical insight levels and that MCT-induced insight improvement will be linked with (compared with controls): ii) reduced symptom severity and iii) lower risk of hospitalizations, iv) lower suicide rates and v) better functioning.

METHODS:

Study design This is a single-center, assessor-blind, parallel group, two-armed randomised controlled trial (RCT) over a 1-year follow-up period. Participants will be assessed at baseline, following which they will be randomised to either group MCT (experimental group) or a support group (control group) and they will be reassessed after treatment, i.e., at approximately 8 weeks, and at 1-year follow-up.

The study protocol has been approved by the local Research Ethics Committee of the Instituto de Investigación Sanitaria (IIS)-Fundación Jiménez Díaz (Madrid, Spain), from which patients will be recruited as detailed below, and registered as RCT-EC044-19_FJD_HRJC.

Recruitment process:

Recruitment will occur at the outpatient clinic known as Centro de Salud Mental de Arganzuela, which is part of Instituto Fundacion Jimenez Diaz, that is, the 'sponsor', at which two consultant psychiatrists and one consultant psychologist will refer potential candidates to the principal investigator of the project (JDLM) to arrange a first pre-recruitment interview during which the relevant information of the project is explained to the candidate in lay terms, including an information leaflet. Those agreeing to participate in the study are screened against the above inclusion/exclusion criteria at this point. Those who are found to fulfill the study selection criteria undertake a neurocognitive assessment. Specifically, the WAIS-IV is administered by a MSc-level clinical psychologist in order to rule out an IQ≤70, which is an exclusion criterion. Those with an IQ>70 are administered the Trail Making Test (TMT), which assesses executive function and has been found to correlate with insight scores in psychosis patients, shortly after the WAIS-IV in order to complete the baseline neurocognitive assessment at this point.

After a short break of around 5 minutes, the 'assessor' (JDLM) performs the MINI International Psychiatric Interview to confirm the diagnosis of psychosis and conducts the psychopathological, insight, metacognitive and functioning assessments detailed below. Also, a set of demographic and clinical variables, which are listed below, are collected at this baseline interview.

All participants will be encouraged to continue taking medication as prescribed by the treating consultant psychiatrist, usually antipsychotics, and to receive non-metacognitive oriented psychotherapy, such as psychoeducation. There will be no restrictions on medication changes over the trial, which can be made by the treating consultant at any time, although medication-related variables will be considered in secondary analyses as potential mediators/confounders. Participants are informed of their right to drop out of the study at any time without having to disclose a specific reason, which will have no implications on treatment or service provision. Participants are not compensated with any amount of money for completing the assessments and/or receiving the interventions, namely MCT or attending a weekly support group.

Randomisation and assessor-blindness:

After the baseline (Time 0) assessment participants are randomised to either MCT or the support group, both of which are run by two researchers, through a computerized plan (no stratification factors) in blocks of 10 subjects (maximum number of each group) and assessor (JDLM)-patient blind. Only one co-principal investigator of the project (EBG) has access to the randomisation plan, but the assessor (JDLM) is not informed of the patient's allocation group, thus ensuring assessor-blindness. Since the patient may find out what intervention he/she is receiving the study cannot be considered to be 'double-blind', although it is 'single-blind', that is, 'assessor-blind'. Also, big efforts will be made on the ground to reduce the risk that the assessor (JDLM) may become accidentally unblinded and patients are reminded that they should not disclose their group assignment at any time. At the end of the baseline assessment, participants receive an envelope with the intervention assignment from an independent administrator who is not involved in the research team. A reminder mobile text-message is also sent to participants within the next 24 hours with the group details (date, time and venue), which is also re-sent again during the 24 hours prior to the appointment, which is very similar to our routine clinical practice. Hence, assessor-blindness is guaranteed at all times throughout the study period since the assessor (JDLM) is not involved in the randomization plan or the active interventions.

Assessments:

Participants will be assessed at three timepoints: i) at baseline (Time 0); ii) after treatment (Time 1); iii) at 1-year follow-up (Time 2). Data on different variables will be collected at each of these assessments.

The assessor (JDLM) completed a rater-training workshop and received individual training in which they were observed as they assessed pilot patients by using the project scales explained below under the supervision of a senior consultant psychiatrist with expertise in the use of these instruments (EBG).

Additional variables:

Demographic and clinical data:

At baseline data on the following demographic and clinical variables will be collected: gender, age at the study inception, nationality, ethnicity, marital status, education level, living status, employment status, ICD-10 diagnosis, duration of untreated psychosis, number of previous admissions, number, date and method of previous suicidal acts, current medications, alcohol/illicit drugs dependence (present/absent), medical comorbidities (present/absent) family history of mental illness (present/absent).

Marital status, living status and employment status will be reassessed at 1-year follow-up as measures of functioning.

Premorbid adjustment:

Premorbid adjustment, which can be defined as 'the degree of achievement of developmental goals' will be retrospectively rated with the Premorbid Adjustment Scale (PAS). Specifically, the PAS provides scores on the level of adjustment over i) childhood (to age 11), ii) early adolescence (age 11-15), iii) late adolescence (age 15-17) and iv) adulthood (age ≥18). With regard to childhood and adolescence, items inquiry about sociability and social withdrawal, peer relationships, scholastic performance, adaptation to school and ability to form socio-sexual relationships. The questions regarding adulthood focus on social relationships by asking about educational achievement, social relationships and level of interest in, and enjoyment of, major life activities, such as work or family.

Neurocognition:

The neurocognitive assessment battery, which will be administered at the three assessments, will include the short version of the Wechsler Adult Intelligence Scale (WAIS) Revised, which estimates current intelligence quotient (IQ), and the Trail Making Test (TMT), which assesses executive function. Subtracting time in seconds to complete TMT task B minus time in seconds to complete TMT task A gives a measure of executive function, whilst controlling for processing speed.

Metacognition:

Three metacognitive tasks will be completed by participants at baseline, after treatment and at 1-year follow-up:

• Jumping to Conclusions (JTC) will be determined with the beads task. During this task, the individual is asked to decide the jar to which the extracted bead belongs on the basis of probability (in task 1 the probability is 85:15, while in task 2 the probability is 60:40). 'JTC' is considered as making a decision after extracting one or two beads.
• The Hinting Task will be used to measure Theory of Mind (ToM) performance. Learning is avoided by using different stories at the three assessments. Cronbach's α was good (0.64) for the Spanish version.
• In addition, the Emotional Recognition Test Faces, which is composed of 20 different pictures representing people's emotions, will be used to evaluate ToM.

Statistical analysis:

Participants' data will be analysed at the end of the study on an intention-to-treat (ITT) basis, thus including all patients with baseline information available. Imputation methods will be used to estimate missing values. All the analyses will be performed with the Statistical Package for Social Science version 25.0 (SPSS Inc., Chicago, IL, USA).

First, demographic and clinical characteristics of groups (MCT and controls), including insight measures, will be compared at baseline so parametric and non-parametric tests will be used as appropriate. Also, for descriptive purposes, between-arm completion rates differences will be analysed at the end of the study.

Second, in order to investigate the primary outcome of the study, namely insight changes over the trial period, univariate analysis of covariance (ANCOVA) models will examine between-group differences (MCT and controls) in the SAI-E and BCIS total and subtotal scores changes from Time 0 to Time 1 and from Time 0 to Time 2 (as the dependent variable), whilst adjusting for baseline data. Hence, the effects of treatment (independent of time) and treatment group allocation*time interactions on insight changes will be investigated. The investigators will also explore within-group scores changes between Time 0, Time 1 and Time 2. Specifically, effect sizes (Cohen's d) and the corresponding 95% confidence intervals (CI) will be estimated from the imputed datasets for between- and within-group insight scores changes. In accordance with Cohen's conventions, effect sizes will be classified as 'small' (d<0.2), 'medium' (d=0.2-0.5) or 'large' (d>0.8). Due to multiple testing and the subsequent risk of type I error, results will be adjusted using Bonferroni correction.

Third, for the secondary outcomes evaluated with continuous variables, i.e., symptom severity (PANSS and CDSS) and functioning (GAF, SDLS and WHODAS), analogous ANCOVA models will be used. For those binary secondary outcomes, namely suicidal behaviour and readmissions, survival analyses, i.e., multivariable Cox regression models, will model time to the outcome event (i.e., first suicidal event or hospital admission, respectively) or the censoring date as appropriate, whilst adjusting for baseline variables.

Finally, variables inter-relationships over the trial period will be examined by means of path analysis through structural equation modeling, thus testing the effect of putative mediators/moderators/confounders/covariates, including neurocognition measures, on the above associations.

Power calculations and estimation of sample size:

Given that the mean SAI-E score for psychosis patients is 13/28 with a standard deviation of around 6, a difference of 2 points (e.g., 13 vs. 15) between groups (e.g., MCT vs. controls), which is considered to be clinically meaningful, is equivalent to an effect size of 0.33 with a two-tailed alpha significant level set at 5%. Under these assumptions, for reaching a sufficient statistical power of β=80% at the end of the study, n=63 subjects in each arm will be needed, that is, a total sample size of N=126 patients, who will be analysed at the end of the study on an intention-to-treat (ITT) basis. Attrition rates in previous RCTs investigating MCT effects on symptoms were low (approximately 10%). However, since participants will be followed-up over a more prolonged period (1 year) the investigators have conservatively assumed a much higher drop-out rate of 50%. Under this assumption, the study will be carried out with an initial sample size of N=252 patients, i.e., n=126 participants in each group/arm at the study inception, which will also allow both ITT and 'per protocol' analyses with sufficient power.

Inclusion Criteria:

• Age: 18-64 years, both inclusive
• Diagnosis: schizophrenia spectrum disorder (F20-F29, ICD-10)
• Outpatient status

Exclusion Criteria:

• IQ≤70 as measured by the short form of the Wechsler Adults Intelligence Scale (WAIS) (Wechsler, 1981).
• History of head injury and/or a neurological condition.
• Having received a metacognitively oriented therapy within the previous 12 months.
• Low level of Spanish.
• Lack of cooperativeness with the assessment and/or "intervention".

Individual participant data will be made available to other researchers, provided the dataset access rules detailed below are complied with.

At the end of the baseline data collection, that is, by April 2020, baseline data will become available to other researchers, provided the access criteria below are complied with.

• An application needs to be made by an independent researcher which will have to specify:
• Variables on which data are requested need to be listed.
• Inclusion/exclusion criteria of participants.
• Brief summary of the study protocol (e.g. specific outcome measures).
• Statistical analysis plan
• Results publication plan. Prior to submitting a manuscript for publication including findings from our dataset, this manuscript will have to be approved, including the relevant email, by the principal investigator of this project, (JDLM).
• Ethical approval from the applicant's local Research Ethics Committee.
• Ethical approval from our local Research Ethics Committee.

After the baseline (Time 0) assessment participants are randomised to either MCT or the control group through a computerized plan (no stratification factors) in blocks of 10 subjects (maximum number of each group) and assessor-patient blind. Only one co-principal investigator of the project (EBG) has access to the randomisation plan, but the assessor (JDLM) is not informed of the patient's allocation arm. Also, big efforts will be made on the ground to reduce the risk that the assessor (JDLM) may become accidentally unblinded and patients are reminded that they should not disclose their group assignment at any time. At the end of the baseline assessment, participants receive an envelope with the intervention assignment from an independent administrator who is not involved in the research team. Hence, assessor-blindness is guaranteed at all times throughout the study period since the assessor (JDLM) is not involved in the randomization plan or the active interventions.