Health institutes call for psychosocial interventions and recovery-oriented approaches as supplement to pharmacological treatment for mental health disorders. Participatory art interventions have been suggested to be promising in promoting recovery by stimulating connectedness, hope, renegotiation of identity, participatory meaning-making and empowerment. Moreover, cognitive literature studies suggest there might be potential benefits of engaging with literature in terms of improved cognition and social cognition. In spite of promising findings, the evidence base is still thin. We have developed REWRITALIZE (REWR), a manualised, recovery-oriented fifteen-session participatory creative writing group intervention, led by a professional author and attended by a mental health professional. The intervention comprises introduction to literary forms, spontaneous writing on those forms, sharing texts and engaging in reflective discussions about them. It is designed to provide a holding and non-stigmatising environment.

The aim of the present study is to evaluate REWR for persons with severe mental illness. This study is a randomised controlled clinical trial (RCT) with an embedded pilot RCT focusing on clinical and personal recovery.

This study is an investigator-initiated, randomised, two-arm, single-blinded, multi-center, waiting list trial. Participants (n=266) with severe mental illness (>18 yrs.) will be recruited at six psychiatric centres in region Zealand and randomised to active (creative writing group + treatment as usual) or control (waiting list + treatment as usual) condition. Assessments will be collected pre- and post-intervention and six months after end of intervention. The primary outcome measure will be the questionnaire of the process of recovery administered at the end of the intervention. Secondary outcome measures comprise measures of recovery, self-efficacy and mentalising assessed at the end of the intervention and six months after the intervention ends. The post-intervention measures will be compared between active and control groups by means of independent sample t-tests. The pilot RCT will focus on a subset of participants (n=70) with schizophrenia spectrum disorders (18-35 yrs), evaluating exploratory outcome measures related perspective-taking, social cognition, cognitive function, psychosocial functioning, and symptom level.

Objective: The study has three objectives. The primary objective is to determine if REWR in addition to standard mental healthcare is more effective than standard mental healthcare alone for promoting personal recovery. The secondary objective is to investigate if REWR in addition to standard mental healthcare is more effective than standard mental healthcare alone in promoting a. mentalising, b. self-efficacy, c. functioning and quality of life. The third objective is to explore if REWR in addition to standard mental healthcare furthers clinical recovery, namely cognition, social cognition, psychosocial functioning, and symptom reduction in a sample of young people with schizophrenia spectrum disorders.

Hypotheses: The main hypothesis is that people living with a schizophrenia spectrum disorder who participate in REWR as an add-on to standard mental healthcare, as compared to those that receive standard mental healthcare alone, will show a greater improvement in personal recovery, measured by the Questionnaire for the process of recovery (QPR) immediately after the end of the intervention (approximately 4.5 months post-baseline). Secondary hypotheses include that those who receive REWR in addition to standard mental healthcare will show a greater improvement in personal recovery, as well as mentalising, self-efficacy, clinical recovery and wellbeing immediately after the end of the intervention (approximately 4.5 months post-baseline), as well as six months later (approximately 10.5 months post-baseline).

Ethical considerations: Information about the studies is presented to all potential participants both verbally and in written form so they can make an informed decision about their participation before signing the informed consent forms. Research assistants will make clear to the candidates that participation is voluntary, and that withdrawal can occur at any time without consequence for treatment possibilities. Decisions regarding participation will not influence clinical care in any way. At the information meeting and in the informed consent form it will thus be made clear that the intervention is voluntary and safe, and that participants at any time can stop their participation in the project and withdraw their consent without any consequences for the standard treatment they receive. If the participants wish to withdraw from the experiment during the course, there are three options: 1) They withdraw from the intervention but not from the study, therefore they will participate in the data collection at the follow-ups. 2) They withdraw from both the intervention and the study, but data from the baseline measures can be used in the research project. 3) They withdraw from both the intervention and the study, and their data will be deleted.

The study will be conducted in agreement with the Declaration of Helsinki. The Regional ethical committee for the Capital Region has decided that the study is not required to seek approval under their legislation, which covers human participation in medical trials. We will adhere to all other ethical standards guiding good practice in research. The study has been registered with the Danish Data Protection Agency (p-2023-14655), and we will comply with all regulations regarding data security, and make sure that all technical solutions comply with GDPR regulations. The study will be registered at clinicaltrials.gov. If any modifications are made to the data collection or data analyses, these protocol amendments will be reported to the Danish Data Protection Agency and to clinicaltrials.gov.

The intervention is not expected to have any serious side-effects. This is supported by the pilot study and similar studies. The social setting can be associated with anxiety and discomfort, at least initially, as known from previous group start-ups. A mental health care professional will always be present in the group (the co-conductor, see REWRITALIZE under Interventions) and this is expected to reduce participants' distress and to regulate reactions including crisis following the group activity. Supportive sessions are offered by the co-conductor if needed.

Participant Timeline: Clinical staff will inform potential candidates about the study, and persons interested in participating in the study will contact a research assistant via e-mail and arrange for a meeting. The research assistant meets with the potential participant online and provides further information about the study. If the participant consents to participate, they will receive a consent form via encrypted personal mail. When the participant has provied written consent, the research assistant informs a research clinician who confirms diagnosis of participant via their health record. Then the research assistent send out an e-mail to the participant with link to questionnaires comprising the outcome measures. When the participant has filled out the questionnaires, the research assistant assign the participant to either the intervention group or the control via the randomisation module in REDcap. The research assistent informs the participant of the result of the randomisation. The research assistant also informs the co-therapist in the REWR intervention about the group allocation, and the co-therapist invites the participants in the intervention group to an introductory meeting prior to the REWR intervention. The creative writing group will start when 8-10 participants have been allocated to the active condition.

Participants will be followed up immediately after the end of the intervention, corresponding to approximately 4.5 months after baseline and six months later, approximately 10.5 months after baseline.

For the pilot RCT 70 participants age 18-35 with schizophrenia spectrum disorder will be invited to participate in a extended assessment by a research clinician at baseline and follow-up at 4.5 months. If the participant consents, a meeting will be arranged at the study site for the additional assessment tools. A research clinician will validate the diagnosis of schizophrenia spectrum disorder via the present state examination and measure cognitive performance, social cognitive performance, psychosocial functioning, and positive and negative symptoms. Outcome measures include the clinician-rated PANSS-6 and GAF-F (see Outcome measures), three tasks (SCIP-D, TASIT and VISPT, see Outcome measures), and a link will be provided for the self-report questionnaires. Participants may fill out the questionnaires either at home, or with the help of a research assistant at site using the link. The data collection, including questionnaires, interviews and tasks, takes some time. The participants will be given the opportunity for breaks and reflection time in connection with the interviews. When all baseline data have been collected, randomisation is performed.

Assignment of interventions: After baseline data are obtained, the participants are randomly allocated to either the control (standard mental healthcare) or active group (REWR + standard mental healthcare) with a 1:1 allocation using the randomisation module in REDCap (Research Electronic Data Capture). The randomisation sequence will be generated by a researcher at Copenhagen Research Centre for Mental Health (CORE), who is not part of the present research team, and will be entered into RedCap. The randomisation is stratified by sites. Varying block sizes, unknown to the research team, are used. To ensure concealment, the randomisation schedule is stored away from the research team and the block sizes are not disclosed. The allocation of active or control condition is performed by one (unblinded) research assistant, who informs firstly the participant about the result, secondly the co-conductor at the site who keeps track of what participants have been allocated to active condition. Participants randomised to the control group will be set on a waiting list and will be informed that they will be offered to participate in a creative writing group within one year of the randomisation.

Sample size and power calculations:The minimum sample size is calculated based on the ability to detect a minimal but clinically significant difference between the active intervention group and the control group in the primary personal recovery QPR measure. The minimal clinically significant difference between the study groups has been estimated to 4 points. The assumed standard deviation in the study population of 10, based on trials in similar populations, hence corresponds to an effect size of Cohen's d=0.4. To achieve a statistical power of 90 % at a significance level of 5 %, a total of 266 participants must be included in this study to detect a difference; 133 in each group. Based on the needed participants, power calculations for secondary outcomes were calculated and are shown in table 7. Because of the relatively high drop-out rate in the pilot study (see appendix 2) and similar studies, we will plan for a sample size of 150 participants in the active group and 150 participants in the control group. This should hence allow for at least a 90% power of finding a minimally clinically significant difference in the primary measure, i.e., the QPR measure of personal recovery.

No sample size calculation is required for a pilot RCT, but we estimate 70 participants to be sufficient to investigate the study objective.

Data analyses:The primary outcome measure is personal recovery measured by QPR. To test the research hypothesis, the differences between the active intervention group and the control group will be analysed using independent-samples t-test. Effect sizes to judge clinical relevance will be calculated by Cohen's d. All variables are continous and secondary measures will be tested by the same means. The significance level is set to 0.05. Data analyses will be based on the intention-to-treat principle. Data from all participants will be included in the analysis. In accordance with the intention-to-treat principle, missing data from the follow-up will be imputed using multiple imputation.

The prerequisite for the use of multiple imputations is that data are missing at random or completely at random as opposed to non-ignorable nonresponse. This distinction is important since multiple imputations are based on a statistical estimation of non-existent responses, and the prerequisites for this estimation must be met for the analyses to be valid. Significant prognostic characteristics of the non-follow-up participants will thus be compared with those for whom follow-up data have been collected. Variables where there is a difference between participants and non-participants will be included as co-variables in the analyses of differences between active and control group to make these analyses valid. As a supplementary analysis, we will carry out per-protocol where only cases with follow-up data are included.

A detailed statistical analysis plan will be prepared before initiating analysis and uploaded at clinicaltrials.gov.

Inclusion Criteria:

• Having a severe mental illness
• Minimum 18 yrs. of age
• Master Danish

Exclusion criteria:

• Aggressive or disruptive behaviour that hinders participation

Data sharing is not permissible due to the EU General Data protection Regulation 2016679 (GDPR). Data collected in the study can be made available to other researchers upon reasonable request.

Owing to the nature of the intervention, neither participants nor staff can be blinded to allocation. They will be instructed not to disclose the allocation status to the researchers. Baseline data will be collected prior to randomisation and the research clinicians and research assistant that collect follow-up data will be blinded. If he or she is unblinded during the follow-up assessments, for instance if the participant accidentally reveals his or her allocation status, the participant will be invited to another follow-up session, with a blinded assessor. An employee outside the research team will extract data from REDCap on study completion and group allocation will be coded with A and B to ensure blinding of the researchers while analysing data, interpretation of data, drawing conclusions, and writing up reports.